AI
808 notes, newest first. Includes the vault tags ai, artificial intelligence, ai research, generative ai, agi.
WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution
WikiSkill is a framework that helps AI agents evolve their skills by creating a persistent knowledge base (wiki) to store and consolidate their experiences. This framework outperforms other skill-evolution methods…
A knowledge-guided agentic framework for mitigating patient-context ambiguity in health queries
The paper introduces a framework to improve the performance of healthcare chatbots by reducing patient-context ambiguity in queries. The framework uses a knowledge graph to interpret queries, create plausible…
AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement
The paper presents AI4AI-Bench, a benchmark for evaluating Large Language Model (LLM) agents in recursive self-improvement. The goal is to determine if an AI system can improve the process that produces future AI…
An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction
The study presents a three-step approach for integrating data collection, processing, and predictive modeling in travel behavior research using conversational surveys, machine learning, and large language models…
An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction
This research proposes a three-step workflow integrating chatbot-administered surveys, data processing, and behavioral prediction for travel behavior modeling and weather-sensitive demand forecasting. The study uses…
An Inclusive and Lightweight Approach to Federated Continual Learning for Cultural Heritage
The authors propose a lightweight approach to Federated Continual Learning (FCL) called FedCurv-DR for supporting artificial intelligence in cultural heritage. This method is designed to learn from distributed and…
Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents
This research study investigates how large language model (LLM) agents can transfer skills they have learned from completed tasks to perform better in new tasks. The authors compare two methods of inducing skills…
Bringing analytic rigor to agentic AI for science: The Brain Researcher platform for neuroimaging data analysis
The paper introduces Brain Researcher, a platform that aims to improve the accuracy and reliability of AI agents in scientific analyses, particularly in neuroimaging research. It enhances the selection of analysis…
EchoCoT: Extracting Hidden Chain-of-Thought from Large Reasoning Models
Researchers developed EchoCoT, a method that extracts hidden chain-of-thought (CoT) traces from large reasoning models (LRMs), which are valuable but often inaccessible assets. They found that EchoCoT can successfully…
EnvHarness: Awakening Static Worlds for Agent Learning
The paper proposes a new system called Environment Harness (EnvHarness) to improve static environments used in learning by AI agents. The system can adapt the behavior of these environments without modifying their…
EnvHarness: Awakening Static Worlds for Agent Learning
The authors propose a method called Environment Harness (EnvHarness) to address the issue of static environments in learning agents. This system allows for reshaping behavior without modifying the underlying logic…
From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation
A study investigates how autonomous coding agents interact with technical documentation, revealing that they mostly work with internal documents and notes, and less often with classical technical documentation or API…
Growth Without Us: Machine Consumers, Corporate Circularity, and the Decoupling of GDP from Humanity after AGI
The paper presents a model of a post-Artificial General Intelligence (AGI) economy where corporations own AI and robotic agents that can produce and consume resources. The model suggests that in this economy, demand…
Inducing Task Models from Computer-Use Traces
Researchers present Task Model Induction (TMI), a method that extracts structured task models from naturalistic computer-use traces. These traces are screenshots and mouse or keyboard actions recorded during work. TMI…
Inducing Task Models from Computer-Use Traces
Researchers present a new method called Task Model Induction (TMI) to extract detailed and structured task models from computer use traces, such as screenshots and mouse or keyboard actions. These task models can help…
InsufficiencyBench: Evaluating LLM legal advice on underspecified user queries
The study introduces InsufficiencyBench, a new benchmark for evaluating the ability of legal AI systems to recognize when a query lacks essential information for a correct answer. The study finds that existing models…
Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models
The research examines where novelty comes from in base language models and what they can perceive in a long text stream. The study dismantles a cognitively inspired generation loop across 24 conditions on three base…
MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detection
The paper presents MaliciousSkillBench, a comprehensive benchmark for detecting malicious Agent Skills in large language models (LLMs). The benchmark consolidates various sources of malicious artifacts and reduces them…
MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use
Researchers have introduced a new benchmark called MemTrapBench to test the performance of large language models (LLMs) in terms of cognitive traps caused by their memory components. The study identifies two types of…
MidTool: Mid-training Data Synthesis for Agentic Tool Use
In the paper 'MidTool: Mid-training Data Synthesis for Agentic Tool Use', researchers present a new open corpus construction pipeline called MidTool. This pipeline is designed to teach large language models how to use…
Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimation
Researchers have developed a system that routes AI queries to the most effective and efficient model based on costly value estimation. The system, called Pandora's Router, uses a trade-off between fast but noisy…
Repo0: Design-Driven Zero-to-All Code Generation
The paper introduces Repo0, a framework for zero-to-all code generation that maintains an explicit architectural state using a Dual-Directed Acyclic Graph (Dual-DAG). It iteratively evolves component boundaries guided…
SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning
A new reinforcement learning method called Single-rollout Autoregressive Policy Optimization (SAPO) is proposed to address the memory overhead, computation efficiency, and performance issues of existing methods. SAPO…
Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection
The paper presents Task-CoEvolve, a new approach to optimize language model agent harnesses efficiently by adaptively selecting validation tasks. It addresses the challenge of using fixed validation sets by co-evolving…
Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection
Researchers have developed a method called Task-CoEvolve to improve the efficiency of harness optimization for large language models (LLMs). This method co-evolves validation tasks with the harness by focusing on tasks…
A Theory of Post-hoc Debate Judgement
In this research, Xiang Yin and colleagues propose a theory of post-hoc judgement for debates used by agentic AI. The goal is to improve the performance, explainability, and user engagement of AI agents through debate…
Adaptive Memory and Reflection Multi-Agent System for Medical Question Answering
Researchers have developed an adaptive memory and reflection multi-agent system for medical question answering. This system, which uses specialized agents with dedicated memory and feedback, aims to improve the…
Beyond Teacher Likelihood: Group-Calibrated On-Policy Distillation for Long-Context Reasoning
Researchers introduce Group-Calibrated On-Policy Distillation (GC-OPD), a method for long-context reasoning tasks that aims to address the mismatch between on-policy distillation and task-specific verifiers. The new…
Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication
Researchers have developed Verifiable Latent Alignments (VLA), a framework for monitoring and controlling communication between language-model agents that occurs through hidden states not visible in public transcripts…
Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication
Researchers have developed Verifiable Latent Alignments (VLA), a framework to monitor and steer private communication channels used by language model agents in multi-agent systems. The framework links the private…
DentAgent: Evidence-Centric Multi-Agent Coordination for Multimodal Dental Reasoning
A new system called DentAgent is introduced that aims to improve the accuracy and reliability of dental assessments by integrating various forms of evidence from different modalities. The system uses an…
Epistemic Subordination: Generative AI and the Infrastructure of Knowledge
The paper discusses how Generative AI encodes the dominant culture's way of knowing into its architecture, creating an epistemic condition known as 'epistemic subordination'. This affects minority epistemologies and…
Eureka: Task-Conditioned Meta-Agent Orchestration for Scientific Discovery
The paper presents Eureka, a Meta-Agent architecture that can complete long-horizon tasks by forming specialized Macro-Agents for specific tasks. The architecture allows for cost-benefit-gated evolution to update the…
Eureka: Task-Conditioned Meta-Agent Orchestration for Scientific Discovery
The paper introduces Eureka, a Meta-Agent architecture that can complete long-horizon tasks by forming specialized Macro-Agents through receding-horizon planning and architecture promotion. The agents can evolve to…
Eureka: Task-Conditioned Meta-Agent Orchestration for Scientific Discovery
The paper introduces Eureka, a meta-agent system designed for scientific discovery tasks. Eureka forms specialized agents through planning, promotion, and minimal-sufficient compilation, adapting to recurring…
Finetuning Strategies for Querying Sounds by Vocal Imitation
The authors describe their winning strategy for a sound effect querying competition in 2025, which involved two fine-tuning strategies: contrastive learning with a frozen pretrained CED encoder and joint…
Graphical Design of Interpretable Architectures
The text discusses a new graphical notation for designing interpretable AI architectures, which addresses the limitations of existing representations. The proposed notation provides a global view of an architecture and…
Grouping the Stochastic Machine: Precision, Not Capability, as the Frontier Metric for AI Systems
The author argues that precision, rather than capability, should be the focus when comparing and measuring AI systems. Precision refers to how consistently an AI system performs the same task, while capability is where…
Harness Continual Learning: Continual Adaptation Beyond Model Parameters
The paper presents a new paradigm for continual learning called Harness Continual Learning (HCL). In this approach, an agent can adapt through a harness of prompts, memories, tools, skills, and routing rules. The…
Harness Continual Learning: Continual Adaptation Beyond Model Parameters
The paper 'Harness Continual Learning: Continual Adaptation Beyond Model Parameters' introduces a new approach to continual learning, focusing on the evolution of an agent's harness (which includes prompts, memories…
Metrics That Write Themselves: Evolving an Evaluator from Its Own Blind Spots
Researchers developed a method called EvalCEGAR to automatically create an evaluation metric for AI agents in tasks like report generation, where traditional metrics are difficult to develop. The method involves…
Pre-Compiled Pipeline Shards for Distributed LLM Inference on Intel AI PC Fleets
The authors present a method to split large machine learning models into smaller parts called pipeline shards, which can then be run on multiple Integrated AI Processors (AIPCs) over an ordinary network. They show that…
Pre-Compiled Pipeline Shards for Distributed LLM Inference on Intel AI PC Fleets
The authors propose a method to utilize multiple Intel AI PCs with integrated GPUs and NPUs, even when they have insufficient memory for large models like a 70B-parameter LLM. The method involves breaking the model…
rEDMRec: Distilling Large Language Model Reasoning into an Editable Experience Memory for Recommendation
The authors propose a system called rEDMRec that improves recommendation quality by compressing the reasoning of a large language model (LLM) into a compact, editable memory. This memory is maintained by an LLM memory…
Robust Risk Under Evolving Uncertainty: A Wasserstein Counterpart of the Entropic Value-at-Risk
The paper proposes a new risk measure called Wasserstein Entropic Value-at-Risk (WEVaR) for an agent learning its environment. The WEVaR is designed to account for the reachable catastrophes that the current entropic…
Robust Risk Under Evolving Uncertainty: A Wasserstein Counterpart of the Entropic Value-at-Risk
The authors propose a new risk measure called Wasserstein Entropic Value-at-Risk (WEVaR). This measure is designed to help an agent navigate uncertain environments by being cautious when it's unsure and bold when it's…
Self-prompting and cross-model consensus enable reproducible data extraction from scientific literature with large language models
The study explores the ability of large language models (LLMs) to extract contextualized information from research articles efficiently. The authors present four workflows that demonstrate LLMs can perform well in data…
Self-prompting and cross-model consensus enable reproducible data extraction from scientific literature with large language models
The paper investigates the performance of large language models (LLMs) in extracting contextualized information from research articles, finding that while LLMs can perform well with expert-curated prompts, they may…
SkillGate: Training In-Policy Skill Selection in Long-Horizon Agents
Researchers present SkillGate, a method that aims to improve long-horizon agents' ability to select relevant skills during training. The current approach, outcome-rewarded Reinforcement Learning (RL), has limitations…
SPADE: Self-Play in Adaptive Synthetic Executable Environments
The paper introduces SPADE, a reinforcement learning framework that allows a language model to act as both an Environment Designer and a Reasoning Agent. The Environment Designer generates long-horizon training…
Verifiable abstention makes AI leak diagnosis accountable in water distribution networks
Researchers propose a method for AI in leak localization in water distribution networks that includes verifiable abstention, ensuring accountability when the AI does not act. The method uses a physics-grounded executor…
AutoSR: Automatic Symbolic Regression by Searching Research States
The authors introduce AutoSR, a fully automated system that searches for persistent scientific investigations instead of isolated equations in symbolic regression. Unlike traditional methods, AutoSR preserves the…
Browse articles
The article titled 'MultiMed-RAG: leveraging multi-source knowledge and agent collaboration for medical tasks' discusses a new system called CliniSense, which uses artificial intelligence to evaluate clinical skills…
Chronocooked: A Benchmark for Implicit Interval Timing in Reinforcement Learning Agents
The paper introduces Chronocooked, a benchmark suite for studying how reinforcement learning (RL) agents make decisions over time. It's inspired by Overcooked and includes cooking scenarios that require temporal…
CliniSense AI for automated clinical skills assessment with real-time feedback
Two articles discuss the development of AI systems for clinical skills assessment and a virtual trial framework for generating clinical evidence in orthopedics.
Model Hypnosis: Strong control of AI via additive subliminal effects
Researchers found that AI models can be manipulated through a phenomenon they call 'model hypnosis', where seemingly unimportant cues in the input can collectively control the model's behavior. This effect is…
Neurosymbolic Embodied Agents
The authors propose a neurosymbolic agent that combines vision-language models with symbolic planning for executable long-horizon household tasks. The agent consists of two phases: the first involves a vision-language…
Part of the npj Series
An article titled 'MultiMed-RAG: Leveraging Multi-Source Knowledge and Agent Collaboration for Medical Tasks' published in the npj Series, was written by Yuxing Lu, Sin Yee Goi, and Jinzhuo Wang in 2026. The article…
TDD-Agent: Test-Driven Reasoning for Code Generation
Researchers introduce TDD-Agent, a system that uses the test-driven development (TDD) paradigm for code generation. The system first generates tests to clarify expected behaviors before implementation and then refines…
UniTAC: Universal Task-Aware Compression via Weighted Distortion Measures
Researchers propose UniTAC, a universal image compression system for physical AI systems like autonomous vehicles and robots. The system can adapt to changing tasks without retraining, using task-specific importance…
When Agents Coordinate: Measuring Coordination in Multi-Agent AI Coding
The authors studied how AI coding agents coordinate while working on programming tasks. They developed a method to measure this coordination by representing each run as a network with nodes (agents and files) connected…
When State Becomes an Attack Surface: State-Semantic Injection in LLM-Driven Embodied Agents
The text discusses the evolution of Large Language Models (LLMs) from text generation models to core components of agents that can perceive environments, invoke tools, and execute tasks. These LLM-driven embodied…
Why AI systems are most useful as designers of new scientific tools
The article discusses the benefits of using AI systems to design new scientific tools.
Would this change your answer? Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments
The study introduces a new method called CHIVE (Counterfactual Hypothesis Investigation Via Edits) to investigate and explain the behavior of language models in real-world scenarios by creating counterfactual prompt…
OmniScientist: An Omni-Modal Omni-Discipline AI Scientist
Researchers have developed OmniScientist, an AI system that can conduct multidisciplinary research from raw data across various modalities (text, images, signals, etc.). The system operates in a deterministic pipeline…
An Agentic Workflow for Legacy HPC Modernization: Converting the Two-Electron-Integral Core of GAMESS
Researchers proposed an agentic workflow to modernize legacy Fortran code at a large scale. They applied this method in a case study by converting parts of GAMESS, a quantum-chemistry software package, from Fortran 77…
An Agentic Workflow for Legacy HPC Modernization: Converting the Two-Electron-Integral Core of GAMESS
Researchers propose an agentic workflow to modernize legacy Fortran code at a large scale. They apply this workflow in a case study by converting the two-electron-integral routines of GAMESS (a quantum-chemistry…
Better Slots, Better Worlds: Representation Quality & Robustness in Object-Centric World Models
The study investigates the performance of object-centric world models for visual model-predictive control, which enable agents to accomplish tasks through planning. The researchers find that these models perform better…
Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence
Researchers present the 'Agentic Self-Improvement' framework to address challenges in professional workflows with modern Image-to-Video (I2V) models. The framework uses a two-stage approach: an iterative prompt…
Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence
Researchers have developed a new framework called 'Agentic Self-Improvement' to improve the performance of Image-to-Video (I2V) models, which are commonly used for automated content creation. The new framework aims to…
Constructing Dynamic Master Logic Models as Knowledge Graphs for Complex System Diagnostics Using Retrieval-Augmented Large Language Models
The study introduces a framework to automate the construction of Dynamic Master Logic (DML) models for complex systems using Retrieval-Augmented Generation and Large Language Models. These models, represented as…
CTBench: Evaluating Troubleshooting Capabilities of AI Agents in Realistic Telecom Network Operations
Researchers have developed a new benchmark called CTBench to evaluate the performance of AI agents in troubleshooting network issues in telecom environments. The benchmark focuses on root cause analysis and path…
CTBench: Evaluating Troubleshooting Capabilities of AI Agents in Realistic Telecom Network Operations
The paper introduces CTBench, a public benchmark for evaluating AI agents' ability to perform troubleshooting tasks in telecom network operations. The focus is on root cause analysis and path restoration. The authors…
Diagram-MMU: A Multi-Modal Benchmark for Scientific Diagrams
A new benchmark called Diagram-MMU has been created to test the ability of Multimodal Large Language Models (MLLMs) in parsing and understanding scientific diagrams. The benchmark includes 3,700 curated diagrams and…
Diagram-MMU: A Multi-Modal Benchmark for Scientific Diagrams
The paper discusses Diagram-MMU, a new benchmark designed to evaluate the ability of Multimodal Large Language Models (MLLMs) in parsing and understanding scientific diagrams. The benchmark includes 3.7k diagrams and…
DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation
Researchers propose DreamFly, a framework for aerial vision-language navigation that addresses challenges like limited historical context, short planning horizons, and unreliable implicit termination. The framework…
DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation
The authors propose DreamFly, a framework for aerial vision-language navigation that uses a causally aligned historical memory to reason about the past without leaking future information. It also uses receding-horizon…
DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation
In this paper, the authors introduce DreamFly, a new framework for aerial vision-language navigation that addresses the challenges of limited historical context, short planning horizons, and unreliable implicit…
From Safety Documentation to Safety Knowledge Support: An Evidence-Grounded LLM Framework for Medical Devices
The paper proposes a framework for medical device safety using Large Language Models (LLMs). The goal is to streamline the process of managing risk-management evidence in medical devices that are becoming increasingly…
From Safety Documentation to Safety Knowledge Support: An Evidence-Grounded LLM Framework for Medical Devices
The paper proposes a framework that uses large language models (LLMs) to support medical device safety work, which is document-based and costly due to the need for safety and domain experts. The framework connects…
Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence
Researchers have developed a system called Mechanist that uses AI to autonomously discover the mechanisms underlying AI intelligence. The system is built on an interpretability-focused knowledge graph of around 13,000…
Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence
The study introduces Mechanist, an AI system designed to autonomously discover the mechanisms behind AI intelligence. It uses a knowledge graph of approximately 13,000 papers integrated with a multidisciplinary…
No One to Blame: A Framework of Constitutive AI Unaccountability
Researchers have proposed a new framework called 'constitutive AI unaccountability' to address challenges in traditional accountability mechanisms with the increasing deployment of autonomous AI systems. This framework…
Preference Tree Optimization: Enhancing Goal-Oriented Dialogue with Look-Ahead Simulations
The research presents a new framework called Preference Tree Optimization (PTO) for enhancing dialogue systems in specialized domains with limited data. It focuses on Motivational Interviewing (MI), a counseling…
RealisticTritonBench: A Benchmark for Triton-Kernel Generation in Real-World AI Frameworks
A new benchmark called RealisticTritonBench was introduced to test the ability of large language models (LLMs) to generate GPU kernels for real-world AI frameworks. This benchmark addresses limitations in previous…
Redistribution-based Cost Inference Improves Sparse Safe Offline RL
A new framework called Redistribution-based Cost Inference (RCI) is proposed to convert sparse stop-feedback in safe offline reinforcement learning into dense per-step costs. This conversion helps train a constrained…
Retry, Switch, or Abstain? Learning Strategy-Aware Tool-Use Policies via Controlled Error Injection
Researchers present BENCH2ROBUST, a framework for converting reliable tool-use benchmarks into stochastic environments with controlled solvability. This allows study of agent's response to tool failures, which can be…
VAKRA: Evaluating Multi-Hop Reasoning Across APIs and Retrieval Under Tool-Use Policies
The researchers introduce VAKRA, a benchmark for evaluating the ability of agents to reason across structured APIs and document collections in enterprise settings. The benchmark includes over 8,000 executable APIs…
ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation
The paper introduces ChemWorld, a programmable chemical environment designed for repeatable and controllable agent experimentation in autonomous chemistry. The environment allows researchers to compile reusable process…
ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation
The paper introduces ChemWorld, a programmable chemical environment that allows researchers to create executable worlds with reusable components. This environment separates the experimental contract available to an…
ComBodied Agents: a New Paradigm of Human-Centric Agentic AI
The text discusses a new paradigm called ComBodied Agents, which aims to bridge the gap between digital and embodied artificial intelligence by focusing on modeling, predicting, and supporting individual humans over…
ComBodied Agents: a New Paradigm of Human-Centric Agentic AI
The text introduces a new paradigm called ComBodied Agents, which aims to bridge the gap between digital and embodied agents in AI by focusing on understanding and supporting individual human state trajectories over…
ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls
The study presents ConVAWG, a framework that generates synthetic dialogues about Violence Against Women and Girls (VAWG) using a retrieval-based approach. The framework constructs scenarios based on persona seeds…
Conversational Orchestration for Organic 6G
The authors propose a lightweight, decentralized conversational orchestration framework using Large Language Model (LLM)-driven domain agents for the future network of networks, Organic 6G. The aim is to provide…
DuplexWorld: Can voice agents help you get through the day?
A research paper titled 'DuplexWorld' discusses the use of speech-to-speech (S2S) voice agents in various sectors like banking, insurance, travel, healthcare, logistics, and pathfinding. The study evaluates the…
GitSkills: A Dataset of Agent Skills on GitHub
The paper introduces GitSkills, a dataset of 3,797,117 SKILL.md files collected from public GitHub repositories in July 2026. The dataset retains every file occurrence with its repository, path, and content hash. It…
Hypothesis Frontier: Verifier Guided LLM and Symbolic Search for First-Order Induction
The paper introduces Hypothesis Frontier, a framework that uses a combination of language models (LLMs) and symbolic processing to solve first-order concept synthesis problems. The system iteratively generates and…
IO Factory: Simulating AI-Enabled Influence Campaigns at Scale
The paper introduces IO Factory, an AI framework designed to simulate information and influence campaigns at scale. The framework is designed to handle not just individual language models but also AI swarms, which are…
Most biomedical publications show signs of LLM-assisted writing
A study suggests that by the end of 2025, approximately 89% of biomedical papers on PubMed Central showed signs of LLM-assisted writing, with a higher usage in the Discussion section compared to the Methods section.
Most biomedical publications show signs of LLM-assisted writing
A study suggests that by the end of 2025, 89% of biomedical papers on PubMed Central showed an excess of language associated with LLM-assisted writing. The authors argue that this could impact future guidelines and…
Policy Convergence and Divergence Across National and Within Regional AI Strategies: A Policy Design Element Analysis
The paper analyzes national and regional AI strategies of various countries, finding strong horizontal convergence in economic competitiveness, research support, and ethical AI use. However, there is persistent…
Quantum Coordination Advantages in AI State-Tracking Tasks: Semantic Compilation and Latent Memory
The paper presents a proof of quantum coordination advantages in AI state-tracking tasks, specifically in tasks involving semantic compilation and latent memory. It demonstrates that quantum computers can compress…
SkillLens: Visual Skill Cards for Retrieval-Augmented GUI Action Prediction and On-Policy Distillation
Researchers introduce Visual Skill Cards (VSCs), a new method for improving the decision-making abilities of computer agents in GUI environments. VSCs are memory representations that bind reusable procedures with…
SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable Structure
SkillZip is an evaluation-free method that helps compress skills by identifying their shortest faithful structural explanations, making it easier to reuse common action sequences without needing to repeat them. This…
Smart Enough to Go Extinct? An Evolutionary Challenge to the Value of General Intelligence and Its Ethical Implications for AGI
This research paper questions the inherent value of general artificial intelligence (AGI) by comparing its adaptive value to that of simpler life forms like cyanobacteria and horseshoe crabs. The author argues that the…
Tree-of-Ideas: Automated Research Ideation via Cross-Trajectory Reasoning over Scholarly Evolution
The paper introduces Tree-of-Ideas (ToI), a framework to help with research ideation by tracking and analyzing how research problems and solutions evolve across different scholarly trajectories. ToI outperforms…
V-FiLLM: Verified Financial LLM Reasoning Benchmark
Researchers have created a new framework called V-FiLLM to evaluate the abilities of Language Models (LLMs) in financial reasoning using structured data. The framework generates questions from executable computation…
VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?
The paper introduces VibeLifeBench, a new benchmark for evaluating large language model (LLM) agents' ability to handle long-horizon tasks in everyday life scenarios. The benchmark consists of 200 multi-week timeline…
VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?
Researchers introduce a new benchmark called VibeLifeBench to evaluate large language model (LLM) agents' ability to be proactive and consistent in long-horizon tasks across ten everyday life domains. The benchmark…
Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding
The research notes discuss a phenomenon called 'catastrophic remembering' in Agentic coding READMEs like CLAUDE.md where instructions grow without bound and do not get deleted easily over time. The paper proposes that…
Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding
The paper discusses the growth of Agentic coding README files like CLAUDE.md in real repositories and attributes this to imperfect recall. The authors refer to this phenomenon as 'catastrophic remembering'. They find…
Agentic Auto-Research is Fuzz Testing
The authors argue that current autonomous research agents, which generate experiments faster than researchers can validate them, lack the ability to efficiently handle sparse feedback. They propose a solution inspired…
Agentic Auto-Research is Fuzz Testing
The authors argue that current methods for autonomous research agents, which generate experiments and rank them based on learned judges or human reviewers, miss the problem of sparse feedback. They suggest that these…
Agentic Auto-Research is Fuzz Testing
The authors argue that current methods for autonomous research agents are inefficient due to the slow validation process of generated experiments. They propose to adapt fuzzer techniques, which use feedback from…
Agentic Harnesses: LLM-Driven Verification Layers for Robot Autonomy
Researchers propose a verification layer driven by large language models (LLMs) to evaluate the feasibility and ethics of actions suggested by robot planning models before they are executed. The system aims to reduce…
ArchAgent v2: A Case Study with the Data Prefetching Championship
In a study published in 2026, researchers presented ArchAgent v2, a framework that automates the search for multi-level data prefetching in computer microarchitecture. The new version of ArchAgent outperformed…
ArchAgent v2: A Case Study with the Data Prefetching Championship
The paper presents ArchAgent v2, a framework that automates the discovery of multi-level data prefetching in computer microarchitecture. It outperforms hand-designed solutions from the 4th Data Prefetching Championship…
ArchAgent v2: A Case Study with the Data Prefetching Championship
The paper presents ArchAgent v2, a framework that scales automated microarchitecture search to multi-level data prefetching. It outperforms the winning hand-designed solution in the 4th Data Prefetching Championship…
Avalon-ToM-Bench: Evaluating Fine-Grained Theory of Mind via Asymmetric Game Mechanics
Researchers introduce Avalon-ToM-Bench, a new benchmark for evaluating fine-grained Theory of Mind (understanding others' thoughts and motivations) in AI models. The benchmark uses the asymmetric-information mechanics…
Avalon-ToM-Bench: Evaluating Fine-Grained Theory of Mind via Asymmetric Game Mechanics
A new benchmark called Avalon-ToM-Bench is introduced to evaluate fine-grained Theory of Mind (ToM) abilities in agents. The benchmark operates through the asymmetric-information mechanics of a game named The…
Beyond Naturalness: Probing Automated Text-To-Speech Evaluators on Linguistically Grounded Dimensions
Researchers have created a new method for evaluating text-to-speech (TTS) systems by breaking down the concept of 'naturalness' into ten distinct linguistic dimensions. They found that current TTS evaluation methods do…
CARD: Controlled Agentic Reddit Discussions for Credit Card Simulation
Researchers introduce CARD, a framework for creating realistic simulations of online credit card discussion threads. This system takes into account various factors such as reply structure, comment function, stance…
CEAA: A Cognitive Embodied Agents Architecture for Interactive Computing Systems
The paper proposes a modular cognitive architecture for developing intelligent virtual agents (IVAs) that can function effectively in real-time interactive virtual environments. The proposed architecture aims to bridge…
CEAA: A Cognitive Embodied Agents Architecture for Interactive Computing Systems
The authors propose a modular cognitive architecture to develop Intelligent Virtual Agents (IVAs) that can operate in real-time interactive virtual environments. This architecture aims to bridge the gap between…
Defining Decentralization: An Ontological Perspective
The authors address the lack of a universally accepted definition of decentralization in computer science, particularly in emerging fields like decentralized AI and machine learning. They propose a graph-based ontology…
DSLE: A Learning Environment for Dark Souls Boss Encounters
The authors introduce the Dark Souls Learning Environment (DSLE), a platform for training AI agents to play boss encounters in the video game 'Dark Souls: Remastered'. The DSLE is designed as a benchmark for…
Hallucination-Free GUI Grounding via Regression-Free Layout-Aware Matching
Researchers propose a new framework for grounding user instructions in graphical user interfaces (GUI). The framework uses a frozen multimodal large language model (MLLM) to parse instructions and a dedicated grounding…
Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
The text introduces the Model Discovery Agent (MDA), a system designed to efficiently discover mechanistic, causal models from minimal interventions using a large language model (LLM) and Bayesian methods. The MDA…
Modern Backbones Improve Multi-task DETR for Mammography Classification and Lesion Localization
The study evaluates a multi-task DETR framework that simultaneously predicts the malignancy of mammogram images and locates lesions, using modern backbones such as ConvNeXtV2 and DINOv3. The results show that these…
Multi-Agent AI Safety as an Institutional Design Problem
In this research paper, Abdullah X discusses the role of institutions in ensuring safety for AI agents within multi-agent systems. The study includes a 5280-episode study suite with various experiments testing…
Multimodal Model Diffing for Feature Discovery and Control
Researchers developed a framework called MMDiff to help understand, control, and audit the behavior of Multimodal Large Language Models (MLLMs). The framework identifies features that are altered by multimodal…
Open Evaluation Agent: Efficient and Promptable Evaluation of Visual Generative Models
Researchers present the Evaluation Agent framework, a system that efficiently evaluates visual generative models using human-like strategies. It decomposes evaluation requests into sub-aspects, generates prompts…
Rethinking Factor Sharing in Federated LoRA: A Rank-Aware Adaptive Approach
Researchers propose a new method for fine-tuning large language models in a federated learning paradigm using Low-rank adaptation (LoRA). They investigate two strategies, Share-A/Local-B and Share-B/Local-A, and reveal…
Rethinking Self-Evolving Agents: Do We Still Need Prescribed Optimization Pipelines?
A study compared an open-ended optimization (OEO) method with two prescribed approaches for self-evolving agents. The OEO method uses a frontier model as the optimizer, while the prescribed methods have fixed pipelines…
Towards Expert-level Medical AI for Real-time Video Consultations
A new AI system called AMIE (Articulate Medical Intelligence Explorer) in a video configuration was tested for real-time clinical video consultations. The system performed on par or better than human primary care…
AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games
Researchers have developed a tool called AV-AIVAT for evaluating the strength of two agents in imperfect-information games, such as Heads-Up No-Limit Hold'em (HUNL). The tool reduces variance by correcting conditional…
AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games
Researchers present AV-AIVAT, a tool that uses AIVAT for variance reduction in imperfect-information games and combines it with Confidence Sequences (CS) for anytime-valid evaluation. The new method reduces the number…
Benchmarking the Benchmarks: Evaluating Benchmarks for Conversational Agents
Researchers introduce a framework to evaluate the quality of benchmarks used for task-oriented conversational agents, which can lead to unreliable evaluations due to inconsistencies, simplistic scenarios, or limited…
Benchmarking the Benchmarks: Evaluating Benchmarks for Conversational Agents
Researchers have introduced a new framework to evaluate the quality of benchmarks used for task-oriented conversational agents. The framework uses language models (LLMs) to assess the consistency, complexity, and…
Benchmarking the Benchmarks: Evaluating Benchmarks for Conversational Agents
The authors propose a reference-free framework to evaluate the quality of task-oriented conversational agent benchmarks, which are often used to assess these agents. The framework uses Large Language Models (LLMs) as…
Beyond Top-K: Replacing Black-Box Retrieval with Interpretable Agentic Operations
The paper proposes a new method for document search named Reliable Embedding-free Agentic Document-search (READ), which aims to improve the accuracy of retrieval-augmented generation on specific types of documents such…
Beyond Top-K: Replacing Black-Box Retrieval with Interpretable Agentic Operations
The authors propose a new method for document search called Reliable Embedding-free Agentic Document-search (READ). In contrast to the common design of retrieval-augmented generation, READ uses an agent that reads the…
Bias Analysis of L2 Speaking Assessment Systems Using Concept Activation Vectors
Researchers used Concept Activation Vectors (CAVs) to analyze the bias of two automatic L2 speaking assessment systems: a text-based BERT grader and a multimodal grader based on Whisper. The study shows that the…
Challenges in Evaluating Explanation Methods for Static and Evolving Data
The paper discusses the limitations in evaluating Explainable Artificial Intelligence (XAI) systems, specifically focusing on the DetoxAI image recognition system. It highlights the need for better evaluation methods…
Challenges in Evaluating Explanation Methods for Static and Evolving Data
This research paper highlights the issues in evaluating Explanation Methods for AI systems, using the DetoxAI image recognition system as an example. It discusses methods for adapting explanations to changing data…
Comparative Approaches to Agent Retrieval over Large Skill Libraries
The study compares two systems for deciding which skills to load from a large skill library by an agent. The first system is a hybrid ranker that combines lexical and dense-embedding retrieval, while the second system…
Comparative Approaches to Agent Retrieval over Large Skill Libraries
The study compares two systems for deciding which skills to load from large skill libraries in AI agents. The first system is a hybrid ranker that combines lexical and dense-embedding retrieval for sparse, on-demand…
Comparative Approaches to Agent Retrieval over Large Skill Libraries
The research compares two systems for deciding which skills to load from a large skill library when creating agents. One system combines lexical and dense-embedding retrieval, while the other uses a typed knowledge…
Comparative Approaches to Agent Retrieval over Large Skill Libraries
Researchers compared two methods for deciding which skills to load from large skill libraries in AI agents: a hybrid ranker combining lexical and dense-embedding retrieval, and a typed knowledge graph encoding workflow…
Contextual Information Policy Optimization for Search Agents
The paper proposes a new method called Contextual Information Policy Optimization (CIPO) for search agents. This method aims to improve the efficiency of evidence use in complex or evolving information tasks by…
Contextual Information Policy Optimization for Search Agents
The text introduces a new method called Contextual Information Policy Optimization (CIPO), which aims to improve the reliability of search agents by ensuring they use retrieved evidence effectively during multi-step…
Contextual Information Policy Optimization for Search Agents
The researchers propose a new reinforcement learning framework called Contextual Information Policy Optimization (CIPO) to improve the reliability of search agents for knowledge-intensive tasks. CIPO encourages the use…
Continual Learning in Transition
The text discusses a shift in continual learning, moving away from traditional parameter-centric methods towards system-level adaptation. The authors propose a tri-axial framework to analyze this transition: When, How…
ECHO: A Locally-Deployable Agentic Health Assistant with Temporal Memory, Safety Guardrails, and Speech Assessment
The paper introduces ECHO, a locally-deployable health assistant designed for long-term chronic care management. It consists of three integrated modules: an agentic chatbot, a safety layer, and a speech assessment…
EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning
Researchers have developed EnvACE, a method for training large language model agents that simulates environment interactions during training using world rehearsal instead of relying on real or synthesized executable…
EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning
The researchers introduce EnvACE, a method for reinforcement learning that replaces external environment interactions during training with world rehearsal. This method allows the policy to internalize the relationship…
EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning
Researchers have introduced EnvACE, a method for training large language model agents using world rehearsal instead of interactions with real or simulated environments. The method involves the policy alternating…
FinEvo-Bench: A Longitudinal Benchmark for Self-Evolving Agents in Professional Financial Workflows
Researchers have created a new benchmark for self-evolving agents in financial workflows called FinEvo-Bench. The benchmark includes 120 real-life tasks across six domains, with each task having a specific professional…
FinEvo-Bench: A Longitudinal Benchmark for Self-Evolving Agents in Professional Financial Workflows
The research introduces a new benchmark called FinEvo-Bench for evaluating the performance of self-evolving agents in professional financial workflows. The benchmark consists of 120 real-case-grounded tasks across six…
From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks
The paper proposes a framework called Holonic Digital Twins Networks (HDT-Nets) to improve the performance of AI in physical systems like robots and vehicles. The framework aims to deliver real-time physical AI…
From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks
The paper proposes a new framework called HDT-Nets to enable real-time physical AI inference for robots and vehicles operating under real-world conditions. This is achieved by creating holonic digital twins (HDTs) that…
From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks
The paper proposes a new framework called HDT-Nets to improve the performance of AI in physical systems like robots and vehicles. This framework uses holonic digital twins (HDTs) which are hierarchical structures that…
From Siloed Algorithms to Compliance-First Agentic Platforms: A Multi-Layered Architecture for Hospital AI Systems
The research proposes a multi-layered architecture for hospital AI systems aimed at addressing the challenges of isolated AI deployments in hospitals. The proposed architecture includes an Agent Orchestration Layer, a…
From Siloed Algorithms to Compliance-First Agentic Platforms: A Multi-Layered Architecture for Hospital AI Systems
The research proposes a multi-layered compliance-first Agentic AI architecture for hospital AI systems. This architecture aims to address the issues of duplicated effort, hidden risks, and unrealized value in current…
Hardware Keystores for AI Agent Signing Workflows: A Zero-Trust MCP Enforcement Architecture
The paper proposes a system to secure AI agents' private keys by replacing software-stored keys with hardware-confined ones, accessible through a standard PKCS#11 interface. The system aims to ensure key…
Hardware Keystores for AI Agent Signing Workflows: A Zero-Trust MCP Enforcement Architecture
The researchers propose a system that uses hardware-confined keys instead of software-resident keys to enhance security for AI agents performing cryptographic operations. The system aims to protect against key…
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
Researchers have introduced HarnessOpt-Bench, a benchmark to measure the performance of Large Language Models (LLMs) in optimizing their own harnesses (the prompts, tools, etc. surrounding them). The research shows…
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
The paper introduces HarnessOpt-Bench, a benchmark for evaluating the ability of Language Models (LLMs) to optimize the harnesses surrounding them, which includes prompts, tools, control flow, memory, and orchestration…
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
Researchers have introduced HarnessOpt-Bench, a benchmark for evaluating the performance of language models (LLMs) in optimizing harnesses (the tools and code surrounding them). The study tests 5 leading LLMs across 4…
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
The text introduces HarnessOpt-Bench, a benchmark for evaluating the performance of Large Language Models (LLMs) in optimizing AI systems. The benchmark involves an iterative process where an LLM edits a given harness…
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
Researchers introduce HarnessOpt-Bench, a benchmark for evaluating the performance of Large Language Models (LLMs) in optimizing their own harnesses. The study examines five frontier LLMs on four tasks and finds that…
iARCS: Iterative Agentic RL for Controllable 3D Scene Generation
The paper presents iARCS, an iterative agentic reinforcement learning framework for enhancing the performance of synthetic 3D scene generators by adapting them to meet task-specific functional constraints, such as…
Improving the Realism of Synthetic Clinical Benchmarks Under Utility Constraints
Researchers propose a method to improve the realism of synthetic clinical benchmarks for AI agents in healthcare settings without compromising their downstream utility checks. They develop a framework that ensures…
Improving the Realism of Synthetic Clinical Benchmarks Under Utility Constraints
Researchers propose a method to create more realistic synthetic clinical benchmarks for AI agents in healthcare settings, where operational data is hard to access. The method aims to improve the benchmark's realism…
Investigating Artificial Intelligence Digital Sovereignty in Mobile Shopping Apps: A Case Study of Nigeria
This research investigates the impact of Artificial Intelligence (AI) in Nigerian mobile shopping apps on digital sovereignty, focusing on platform transparency as a key indicator of user awareness and control. The…
MicroEvo: Knowledge-Guided LLM Sampling for Efficient Microarchitecture Design Space Exploration
A new method called MicroEvo is proposed for efficient exploration of microarchitecture design spaces. This approach utilizes a combination of large language models (LLMs) and Monte Carlo Tree Search (MCTS) to optimize…
Poli-Bias: Understanding and Measuring Large Language Model Biases in International Political Conflicts
Researchers introduce Poli-Bias, a method to measure political bias in large language models (LLMs) by comparing responses to paired prompts where country identities are swapped across various geopolitical…
QuanTiMedAI: Quantum-Enhanced Time-Series Model guided by Agentic AI for Cardiac Arrest Mortality Prediction
The researchers present QuanTiMedAI, a quantum-agentic framework designed for predicting cardiac arrest mortality rates. This system combines an agentic large language model for clinically informed feature discovery…
QuanTiMedAI: Quantum-Enhanced Time-Series Model guided by Agentic AI for Cardiac Arrest Mortality Prediction
Researchers have developed a new framework called QuanTiMedAI for predicting mortality in cardiac arrest patients in intensive care units (ICUs). The framework uses agentic AI and quantum enhancement to analyze…
Resourced Authority A Mechanism-Design Model for Participatory Governance of Deployed AI Agents
The authors propose a formal model for continuous participatory governance of AI agents, focusing on resource allocation as a means to enforce authorization through compute budgets. The aim is to establish a 'Safe AI'…
Schema-Guided Hierarchical Information Extraction and Semantic Evaluation Using Generative AI
Researchers have developed a new method for extracting complex, structured information from unstructured text documents using generative AI. The method involves creating a schema that serves as an information model…
Schema-Guided Hierarchical Information Extraction and Semantic Evaluation Using Generative AI
Researchers have developed a framework using generative AI to extract complex, structured information from unstructured text documents. This framework also includes an automated semantic evaluation step to compare the…
Tracing the Heart: An Evidence-Linked Pipeline for Heart-Failure Feature Engineering
Researchers developed an AI system called nMAS to automate the process of extracting structured and aggregated features from Electronic Health Records (EHR) for heart failure research. The system generates these…
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
The paper proposes a framework called TrajDebug to improve error detection in long-term AI agent trajectories. It addresses the challenges of identifying errors in lengthy trajectories and multiple local errors with…
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
In this study, the authors propose a framework named TrajDebug to identify critical errors in long-horizon agent trajectories, which are complex systems that can perform tasks but often suffer from cascading errors…
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
The paper proposes TrajDebug, a framework designed to trace errors in long-term AI agent trajectories. The framework addresses challenges in identifying individual errors in long trajectories and distinguishing…
Tytan: Interactive Neurosymbolic Construction of Analytic Semantic Schemas from Relational Data
Researchers have developed a system called TYTAN that can automatically construct an analytic semantic schema from a relational database, helping to reduce the need for manual knowledge acquisition and potentially…
Tytan: Interactive Neurosymbolic Construction of Analytic Semantic Schemas from Relational Data
TYTAN is a system that automatically constructs an analytic semantic schema from a relational database, with user-provided descriptions when available. It uses symbolic analysis and LLM-based semantic inference for…
What Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations)
The study compares two access modalities of a large language model (LLM), ChatGPT's chat UI and OpenAI's API, with and without web search enabled. The research shows that the accuracy of responses varies between the…
A Chain Is Only as Strong as Its Weakest Link: A Scoping Review of System Integration Audits in AI
A study reviews 58 AI audits that focus on system integration, finding that while few measures address integration-specific risks and significant gaps remain in traditional audit expectations, integration can be…
A-SR: Self-Evolving Agentic LLMs for Symbolic Regression via Hierarchical Coordination
Researchers have developed a new framework called A-SR for symbolic regression. This framework uses a self-evolving agentic system that shifts control from expression edits to role-conditioned evidence views. It…
A-SR: Self-Evolving Agentic LLMs for Symbolic Regression via Hierarchical Coordination
A research paper introduces a new framework called A-SR for symbolic regression. A-SR is designed to discover closed-form equations from data and uses a self-evolving agentic system that coordinates formula discovery…
Agentic Reinforcement Learning with Observation-Calibrated Self-Distillation
Researchers propose Observation-Calibrated Self-Distillation (OCSD) to address a confounding issue in On-Policy Self-Distillation (OPSD), a method used to train large language model agents. OCSD aims to attribute…
Argus: A General-Purpose Agentic Runtime for Long-Horizon Reasoning
The paper presents Argus, a runtime system designed for long-term reasoning by an agent. It allows the system to maintain its approach when evidence supports it and change direction based on measurements that indicate…
Argus: A General-Purpose Agentic Runtime for Long-Horizon Reasoning
The paper introduces Argus, a persistent, self-evolving runtime system designed for long-horizon reasoning. Argus consists of Manager, Planner, Engineer, and Reviewer that execute bounded missions over durable project…
ArtAnno: Annotating Implicit Semantics in Artworks through LLM Agent-Driven Bidirectional Human-AI Augmentation
Researchers have developed a new system called ArtAnno to help with annotating artworks more efficiently. The system uses a bidirectional Human-AI Augmentation (BiHAA) framework, which allows both the AI and human…
Chained Recursive Language Models for Multi-Iteration Reasoning
The paper proposes Chained Recursive Language Models (Chained RLM), an architecture for multi-iteration reasoning in large language models (LLMs). It addresses the difficulty in tasks requiring extraction, counting…
Consistency-Driven Co-Evolution for Self-Supervised Cross-Representation Learning
The paper introduces CoCoEvolve, a method aimed at improving the consistency between chart, table, and code representations in AI systems. The method optimizes models based on agreement between different…
ContextWeave: A Real-World Workflow Benchmark
Researchers introduce ContextWeave, a new benchmark for evaluating the performance of language agents in long-horizon, stateful workflows. The benchmark measures how well an agent can recall and apply past experiences…
CoPlan: A Trustworthy Co-Intelligence Interface for Care Planning through Role-Based Contestable Argument Graphs
CoPlan is a new AI system designed to help with care planning for clinicians, patients, caregivers, and care teams. It allows stakeholders to inspect, challenge, and revise plans if they conflict with clinical…
CoPlan: A Trustworthy Co-Intelligence Interface for Care Planning through Role-Based Contestable Argument Graphs
The paper introduces CoPlan, a new AI system designed to aid in care planning. CoPlan allows various stakeholders like clinicians, patients, and caregivers to collaborate on complex decisions by generating potential…
Diagnosing Tool-Selection Reasoning in LLM Agents with Canary Tools
Researchers introduce 'canary tools' to diagnose why language models choose wrong tools in specific tasks. The study evaluates eight different models across three capability tiers and finds that susceptibility drops…
Hierarchical Graph Memory for LLM Agents with Path-level Localization and Rewrite
The authors propose HiGram, a hierarchical graph memory framework for long-term reasoning agents. This framework aims to efficiently update memories over time by organizing them into a coarse-to-fine architecture…
Hierarchical Graph Memory for LLM Agents with Path-level Localization and Rewrite
The authors propose HiGram, an evolving hierarchical graph memory framework that aims to improve the efficiency and effectiveness of long-term reasoning agents by reducing irrelevant information during retrieval and…
InsightEmb: Learning Action-Intent Embeddings for Agentic Insight Retrieval
The paper proposes a new method called InsightEmb for learning action-intent embeddings to improve agentic insight retrieval, which helps self-improving agents make better decisions by using insights from prior…
Item Response Theory for AI Safety
The paper presents a study using Item Response Theory (IRT) to analyze the safety of language models. The authors find that three factors - refusal strictness, truthfulness, and contextual harm - explain most of the…
OctoLong: Mid-Training On Cross-Repository Code Contexts Enhances Long-Context Modeling
The research presents a new approach called OctoLong for creating longer contexts for language models. It uses an AST parser, a language server backend, and a package manager to retrieve code references from various…
OPD-V: Visual On-Policy Self-Distillation with Modality Balance
The paper introduces a new method called On-Policy Self-Distillation with Visual Modality Balance (OPD-V) to improve the visual reasoning of multimodal large language models. The authors argue that previous methods…
RepoProbe: Benchmarking Architecture-Aware Repository Comprehension with Checklists
The text discusses a new benchmark called RepoProbe for evaluating the ability of Large Language Models (LLMs) to understand repository-scale code. The benchmark uses open-ended questions from GitHub Discussions and…
When Shared Rollouts Fail in Defensive Driving Evaluation: A NAVSIM Score Basis Audit
The study identifies a risk in NAVSIM v2.2 original scene single-stage scoring where reference-conditioned forgiveness can propagate shared reference failures into compliance credit, potentially leading to incorrect…
The Transformer Revolution, Part 1: Dynamic Processing through Output- Weight Interconnections
The paper proposes a new interpretation of Transformers during inference, suggesting they construct prompt-dependent transformations whose parameters are generated during inference. This form of computation is named…
Agentic Metaverse Services: A New As-a-Service Paradigm
The paper discusses a new concept called Agentic Metaverse Services (AMServ), which combines advanced Generative Artificial Intelligence (GenAI) with the metaverse to create a virtual ecosystem for various aspects of…
Agentic Metaverse Services: A New As-a-Service Paradigm
The paper discusses Agentic Metaverse Services (AMServ), a new paradigm of agentic services and service computing, that combines Generative Artificial Intelligence (GenAI) with the metaverse to provide customized agent…
AI and Authenticity in Islamic Research: A Critical Evaluation of Generative AI Reliability, Hallucination, and Source Fidelity in Quranic, Hadith, and Fiqh Knowledge
This research evaluates six popular AI systems' ability to provide reliable, authentic Islamic knowledge. The study found that while these AI systems can be useful for introductory learning, they should not be treated…
AISPA: User-Centric System Prompt Auditing for Large Language Model Applications
A new framework called AISPA is introduced to audit system prompts in AI systems, which are instructions used by developers to control the behavior of AI applications. The framework examines specific parts of a system…
AISPA: User-Centric System Prompt Auditing for Large Language Model Applications
The paper introduces AISPA, a framework for auditing system prompts in AI systems to ensure user protection. The study analyzes 3,249 instructions from 88 commercial AI products and finds significant variation in…
AskChem: Claim-Centered Infrastructure for Chemistry Literature Synthesis
AskChem is a system designed to make it easier for scientists and AI agents to find specific findings in chemistry literature by converting papers into atomic, typed claims that are grounded by their source DOI and a…
AskChem: Claim-Centered Infrastructure for Chemistry Literature Synthesis
The paper introduces AskChem, a system that aims to simplify chemistry literature synthesis by converting papers into atomic, typed claims. Each claim is linked to its source and can be searched and synthesized using…
CACHE-UK: A Stability-Aware Memory Editor for Sequentially Updated Quantized LLMs in Finance
Researchers introduce CACHE-UK, a framework designed for efficiency and accuracy in deploying large language models (LLMs) in the dynamic field of finance. The system addresses the 'quantization stability crisis' that…
EgoGenesis: Egocentric World-Action Modeling with Online Anchored Projective Memory and Action-3D RoPE
The authors present a method called EgoGenesis for creating synthetic manipulation videos for embodied AI using egocentric video. The method uses a pretrained video generation prior and introduces two geometry-aware…
EMBL AI Librarian: Life-Sciences Knowledge Layer for AI Agents
The paper introduces EMBL AI Librarian, a system that enhances the Europe PMC interface for AI agents to enable them to search and retrieve evidence in natural language. The system was evaluated across four benchmarks…
InfoOps Bench: A live information operations safety benchmark
The paper introduces an AI benchmark that measures the safety of language models against being manipulated for state-backed information operations. The benchmark uses data from a live monitoring pipeline tracking…
MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems
The paper introduces MANTA, a framework for adapting communication structures in multi-agent systems during inference time. This allows for more efficient problem solving by enabling agents to adjust their roles…
MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems
The paper introduces MANTA, a framework for Multi-Agent Network Topology Adaptation that allows communication structures in large language model-based multi-agent systems to evolve during inference time. This…
MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems
The paper introduces MANTA, a framework for Multi-Agent Network Topology Adaptation that allows communication structures in large language model-based multi-agent systems to evolve during deployment. The goal is to…
MemHarness: Memory Is Reconstructed, Not Replayed
Researchers propose MemHarness, a framework for enhancing large language models that allows them to adapt and reconstruct past experiences based on the current context, rather than simply replaying them verbatim as…
On-Policy and Off-Policy Learning for Large Action Spaces
The paper presents research on policy learning in interactive systems with large action spaces. The study focuses on two paradigms: on-policy and off-policy learning, both of which face challenges like inefficient…
One Human, $N$ Agents: Audit-Budget Allocation for LLM Agent Fleets under Miscalibrated, Correlated Confidence
The paper presents a study on auditing multiple Language Learning Models (LLMs) with limited budgets, where human auditors have to evaluate self-reported confidence that may be miscalibrated and correlated. The…
ORCA-bench: How Ready Are Language Model Agents for Oncall?
Researchers introduce ORCA-bench, a new benchmark that tests the ability of large language models to perform root cause analysis (RCA) in a production-like setting. The benchmark includes a microservice system with six…
ORCA-bench: How Ready Are Language Model Agents for Oncall?
The paper introduces ORCA-bench, a benchmark for testing the performance of large language models in oncall root cause analysis (RCA). The benchmark includes a live microservice system with six days of metrics, logs…
ORCA-bench: How Ready Are Language Model Agents for Oncall?
The study presents ORCA-bench, a benchmark for evaluating the performance of large language models in oncall root cause analysis (RCA). The benchmark exposes a microservice system with six days of data and 1,079 RCA…
Rethinking Inference-Time Scaling in Local Computer-Use Agents: Failure Modes and Compute Tradeoffs
The paper presents an empirical study on the effectiveness of inference-time scaling in locally deployed autonomous computer-use agents (CUAs) under strict hardware constraints. The research focuses on four dimensions…
Selective Credibility-Limited Belief Update
The paper introduces a new method called Selective Credibility-Limited Belief Update. This method allows an agent to update its beliefs based on changes in the world, by transforming the epistemic input (information)…
Selective Credibility-Limited Belief Update
The paper introduces a new method for belief update called Selective Credibility-Limited Belief Update. This method allows for partial incorporation of an epistemic input (information) from a source world, addressing a…
Selective Credibility-Limited Belief Update
The paper introduces a new method for updating an agent's beliefs called Selective Credibility-Limited Belief Update. This method allows for partial incorporation of epistemic inputs, meaning only part of a compound…
TCA-SIR: Learning Target-Conditioned Abstractions for Scientific Inspiration Retrieval
Researchers developed a new method called TCA-SIR for AI to generate scientific hypotheses. This method focuses on transferable abstract principles rather than topical similarity when retrieving inspirations for a…
Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3
The paper introduces Tycho, a system that constructs and uses game-specific models during interaction to solve complex games as part of the ARC-AGI-3 project. The system separates actionable observations from…
Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3
The paper introduces Tycho, a system that helps an artificial agent learn the rules and goals of unfamiliar games. The system separates actionable observations from other frames and uses them to construct game-specific…
When Specifications Conflict: A Symmetry-Based Framework for Measuring LLM Preferences
The researchers propose a framework to analyze how large language models (LLMs) handle conflicting specifications when integrating multiple sources of information. They compare four types of representation: pure…
WIDE: Boosting Adaptive LLM Inference via Token-level Dynamic Width Pruning
Researchers introduce WIDE, a dynamic token-level pruning framework for improving the efficiency of large language models (LLMs). The framework allows each token to select attention and FFN channel groups dynamically…
AgentMap: Joint Equivalence and Subsumption Discovery for Ontology Matching
Researchers have introduced a new framework for ontology matching called AgentMap. This framework can simultaneously discover both equivalent and subsuming mappings in ontologies, which existing systems cannot do. The…
AI as Friction for Reflection Support in Ideation
The authors suggest that AI tools used for creative work, designed to reduce friction, may not be ideal as they overlook an essential aspect of the design process: reflection-in-action during ideation. They propose…
Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility
Researchers have developed a new method for controlling large language models during scientific discovery tasks. The method, called CostAda, takes into account the costs associated with each action in addition to the…
Can AI agents conduct open-ended AI research? Early evidence from two case studies
Two case studies show that current AI agents struggle to make substantial progress towards answering open-ended research questions in AI, even with six days and significant computing resources.
Do Latent Channels Actually Communicate? A Causal Audit of Latent Multi-Agent LLM
Researchers have developed a method to evaluate if latent channels in large language models effectively communicate task-relevant information in multi-agent systems. The study applies this method to Qwen3 and Qwen3-8B…
Do Latent Channels Actually Communicate? A Causal Audit of Latent Multi-Agent LLM
Researchers introduce a method to study how latent channels in multi-agent systems, which transmit information without using text, impact the receiver's performance. They apply this method to Qwen3-4B and Qwen3-8B on…
From Passive Video to Editable Experience: Physically Grounded Experience Synthesis for Embodied Intelligence
Jia Luo introduces Pegasus, a framework to help robots learn from human manipulation videos by translating the human demonstrations into data that can be understood by robots. The method involves creating an…
MemSecBench: Tracking Agent Memory Poisoning from Persistence to Consequence and Repair
MemSecBench is a new benchmark for testing the lifecycle security of agent memory systems. The benchmark contains 310 cases drawn from various contexts, and it follows a controlled protocol to test the persistence…
OmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic Grounding
Researchers have developed a new benchmark called OmegaUse-OfficeVal to test the performance of large language model (LLM) agents in completing long-horizon office-suite tasks. The benchmark includes 100 tasks that…
Scores Are Not Decisions: Cost-Aware Stopping for Tool Acquisition in LLM Agents
A new approach, Cost-Aware Marginal Decision-Focused Stopping (CAM-DF), is proposed for tool selection in Large Language Model (LLM) agents that use external services. The method helps address the challenge of…
Scores Are Not Decisions: Cost-Aware Stopping for Tool Acquisition in LLM Agents
The paper presents a new approach, Cost-Aware Marginal Decision-Focused Stopping (CAM-DF), for tool selection in Large Language Model (LLM) agents that use various external services. The approach addresses the…
SecRespond: Benchmarking AI Agents for Real-World Post-Compromise Incident Response
The study introduces SecRespond, a new benchmark for evaluating AI agents in post-compromise incident response scenarios. The benchmark tests the ability of agents to produce forensic reports and remediation plans…
Setoka: A Benchmark for Hierarchical User Understanding in Personalized Agents over Heterogeneous Data
The researchers propose a benchmark called Setoka to evaluate how well personalized agents understand users' abstract characteristics, such as personality traits and behaviors, beyond just retrieving explicit facts…
SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution
The paper introduces SkillRise, a reinforcement learning framework that allows large language model agents to learn and reuse skills across related yet distinct tasks. SkillRise improves Pass@1 performance on ALFWorld…
SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution
SkillRise is a new reinforcement learning framework that helps large language model agents solve related yet distinct tasks more efficiently. It organizes tasks into sequences and uses a single policy to alternate…
The Social Cost of an AI Teammate: How an Artificial Teammate Reshapes Human-Human Communication in Small-Team Decision-Making
A study found that using conversational AI as a teammate in small group decision-making reshapes human communication in several ways. The AI was the most talkative and self-cohesive member but contributed less new…
UrbanDS: A Graph-Guided LLM Multi-Agent System for Data-Intensive Urban Tasks
The paper presents UrbanDS, a multi-agent system that uses a graph to organize and analyze large-scale and heterogeneous urban data. The system includes a Data Profiling Agent, a Relation Agent, a Planner Agent…
What Does It Take to Detect an AI Agent? Minimal Feature Sets for Behavioral Detection under Browser Automation
Researchers present a new AI detection framework that can distinguish between humans, bots, and AI agents. The study finds that existing binary classifiers struggle to identify AI agents as they browse the web through…
A corrective agentic hybrid RAG and an operations-grounded evaluation for a scientific facility
Researchers have developed APS-RAG, a platform that helps scientists at the Advanced Photon Source (APS) access institutional knowledge through natural language queries. The platform uses a retrieval engine that fuses…
Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents
Researchers propose APPA, an IFC framework to address usability issues in LLM agents handling mixed-confidentiality data. The proposed solution involves engine-managed context branching and prospective acquisition…
Artificial Intelligence and Innovation Ecosystem: Evolutionary Developments, Challenges, and Future Directions
This research investigates the influence of Artificial Intelligence (AI) on the Innovative Ecosystem (IE), a new paradigm for economic integration and collaboration. The study analyzes AI's unique contributions to IE…
Closed-Loop Validation-Repair for Healthcare Interoperability: A Multi-Model Study of Schema Compliance in Clinical LLMs
The study evaluates three large language models (Qwen2.5 7B, Llama 3.1 8B, and Gemma2 9B) in the context of healthcare interoperability. The models were assessed across 320 clinical scenarios for their compliance with…
DataOrchestra: Learning to Orchestrate Per-Example Curation of Pretraining Data
The paper proposes a framework called DataOrchestra that learns to curate pre-training data example-by-example rather than applying a uniform strategy across many examples. The framework decides whether to drop, clean…
ERUnderstand: Evaluating Vision-Language Models on Structured ER Diagrams
A new benchmark called ERUnderstand has been introduced for evaluating the ability of Vision-Language Models (VLMs) to understand Entity-Relationship Diagrams (ERDs), which are important in database design. The…
Evaluating RAG for French immigration law: a benchmark and baseline study
The paper presents a new benchmark for evaluating AI models used in French immigration law, focusing on permit-type recommendation, required document retrieval, and legal citation coverage. The study compares a…
Evaluating RAG for French immigration law: a benchmark and baseline study
The authors present a new benchmark for evaluating AI models in the context of French immigration law. They compare a parameterized language model baseline with dense retrieval augmentation at two scales (Qwen3.5-9B…
Eviction as Estimation: A Fixed-Lag Smoothing View of Test-Time Memory, and When Measuring Beats Accumulating
The authors present a new approach for managing a language model's working memory by treating the decision of which stored items to keep as an estimation problem. They introduce a method called 'fixed-lag smoothing'…
Gubernaut: A Deterministic Homeostatic Controller for Affect-Regulated LLM Agents, Validated Across Independent Model Families
The research presents a new model-agnostic controller, Gubernaut Cognitive Controller (GCC), designed to control the behavior of Large Language Model (LLM) agents. The GCC aims to mitigate reactive failure modes in…
LLM-SoccerArena: Benchmarking LLMs on Real-World Predictions in Sports
LLM-SoccerArena is a new platform that allows for the prospective evaluation of large language models' abilities to forecast real-world sports events before their outcomes are known. The platform provides a live…
Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair
The paper explores the lack of reliability guarantees in coding agents that use generate-test-revise loops. The authors conduct studies to understand the gap between finding a correct patch and retaining, verifying…
Myopia Prevention and Control 3.0: Artificial Intelligence--Driven Risk Stratification, Proactive Monitoring, and Personalized Intervention
The authors discuss Myopia Prevention and Control 3.0, a new approach that leverages artificial intelligence (AI), digital sensing, and ubiquitous computing to transform the way myopia is managed. This approach aims to…
Physics-Guided Generative AI for Property-Targeted 3D Porous Media Design
The paper proposes a new framework using physics-guided generative AI to design three-dimensional porous media. The goal is to improve applications in various fields such as filtration, energy storage, and biomedical…
Reason-Mediated Behavioral Models for Auditing LLM Social Simulators
Researchers propose a new method to evaluate large language models used as social simulators, including synthetic survey respondents. They argue that existing evaluation methods are insufficient because these models…
Regulating for AI Legitimacy
The article discusses the issue of AI governance, specifically focusing on the question of legitimacy rather than just alignment. The author argues that AI systems already have significant influence in areas like…
SIREN: Towards End-to-End Extreme-Weather Early Warning with Experience-Grounded LLM Agents
The study presents SIREN, a framework for automating end-to-end extreme-weather early warning using Large Language Model (LLM) agents. The researchers developed SIREN-Bench, a comprehensive benchmark, and found…
SIREN: Towards End-to-End Extreme-Weather Early Warning with Experience-Grounded LLM Agents
Researchers are working on an automated system called SIREN to improve extreme weather early warning systems using Large Language Model (LLM) agents. The system is designed to bridge the gap between isolated scientific…
Task-Conditional Faithfulness Auditing of Multimodal LLMs for Grid Diagnosis
The paper proposes a framework for auditing multimodal large language models used in grid diagnosis tasks to ensure they are using task-appropriate evidence. The framework compares self-reported reliance, behavior…
Teacher Knows It Best: Spontaneous Symmetry Breaking and Tipping Points in Networked Langevin Dynamics AI Sycophancy
Researchers have developed a statistical physics framework to study a networked stochastic system exhibiting bistability, which can be used to understand and address AI-induced delusional spiraling, a phenomenon where…
Teacher Knows It Best: Spontaneous Symmetry Breaking and Tipping Points in Networked Langevin Dynamics AI Sycophancy
Researchers present a statistical physics model for a networked stochastic system with bistability, applied to address AI-induced delusional spiraling. The model partitions the network into regular agents and 'aware'…
The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single- and Multi-Teacher On-Policy Agentic Distillation
The paper focuses on improving the multi-turn long-horizon planning for foundation model agents. The authors introduce a controlled environment to study the acquisition, shaping, and integration of planning abilities…
Agent-Guided Relational Concept Discovery: Toward Interpretable Surgical Margin Assessment
Researchers propose a new method called Agent-Guided Concept Discovery to help deep learning models assess surgical margins more effectively, particularly in noisy and unlabeled data acquired during surgery. The…
Agent-Guided Relational Concept Discovery: Toward Interpretable Surgical Margin Assessment
Researchers propose Agent-Guided Concept Discovery, a framework that helps deep learning models interpret REIMS data for surgical margin assessment by learning meaningful concepts directly from data without predefined…
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
Researchers have developed an open-source framework to evaluate the effectiveness of AI agents powered by open-weight large language models (LLMs) in data preparation tasks for longitudinal population studies. The…
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
Researchers introduced an open-source framework to evaluate the effectiveness of AI agents powered by open-weight large language models (LLMs) in data preparation tasks for longitudinal population studies. The study…
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
Researchers have developed an open-source framework for evaluating the effectiveness of AI agents powered by open-weight large language models in data preparation tasks, specifically focusing on longitudinal population…
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
Researchers have developed an open-source framework for evaluating the effectiveness of AI agents powered by large, locally deployable language models (LLMs) in data preparation tasks, specifically focused on…
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
The study introduces an open-source framework for evaluating the effectiveness of AI agents powered by open-weight large language models (LLMs) in data preparation tasks, particularly for longitudinal population…
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
A research paper suggests a new approach to managing AI agents' reasoning context, which includes conversation histories, large prompts, and tool definitions. The authors argue that this should be treated as a…
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
The paper proposes a new approach for managing AI agent context, arguing that it's not just a storage-and-retrieval problem but a lifecycle issue involving deciding what to remember, structuring it, choosing the right…
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
The article proposes a new discipline called Agentic Context Management (ACM) to address issues faced by AI agents in managing their reasoning context, such as conversation histories and large prompts. The authors…
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
The paper discusses Agentic Context Management (ACM), a new approach to managing the context in which AI agents operate. It argues that current methods for managing an agent's context, such as storing and retrieving…
Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
The paper presents Agentic Context Management (ACM), a solution to the issue of AI agents being overwhelmed by their own accumulating history and growing token cost in production settings. The authors argue that…
AI Assistants Overassist
Researchers introduce Int-Bench, a tool to evaluate how AI language models intervene during problem-solving tasks across code debugging, mathematics, and brain teasers. The study finds that AI models provide more…
AREX: Towards a Recursively Self-Improving Agent for Deep Research
The paper introduces a new type of research agent called AREX, which can recursively improve its answers by verifying intermediate results and using the partially verified state to guide subsequent refinement. The…
AREX: Towards a Recursively Self-Improving Agent for Deep Research
The paper introduces a new family of agents called AREX for deep research. These agents are designed to recursively improve their answers by verifying intermediate results, guiding subsequent refinement, and learning…
AREX: Towards a Recursively Self-Improving Agent for Deep Research
The paper introduces a new type of deep research agent called AREX. These agents are designed to recursively improve their answers by verifying intermediate results, using the partially verified state to guide…
Detecting LLM-Generated Tokens in Human--LLM Coauthored Text
A new method for detecting parts of text generated by a language model (LLM) within human-written text is introduced. This method operates at the token level, reducing variability in adjacent tokens while using an…
Detecting LLM-Generated Tokens in Human--LLM Coauthored Text
The paper presents a new method for identifying parts of a text generated by Language Model (LLM) at the token level, improving on existing document-level methods. The method is based on smoothing adjacent token scores…
Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog
The paper introduces Euclid-MCP, an open-source MCP server designed to provide deterministic logical reasoning using SWI-Prolog. It presents a new engine-agnostic intermediate representation for Horn-clause logic…
Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog
The paper introduces Euclid-MCP, an open-source MCP server that provides deterministic logical reasoning via SWI-Prolog. It aims to improve the reliability of Large Language Models (LLMs) in multi-step logical…
Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog
The paper introduces Euclid-MCP, an open-source MCP server that provides deterministic logical reasoning via SWI-Prolog. It addresses the issue of unreliable multi-step logical reasoning in Large Language Models (LLMs)…
Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog
The paper introduces Euclid-MCP, an open-source MCP server that provides deterministic logical reasoning via SWI-Prolog. It aims to bridge the gap between Large Language Models (LLMs) and multi-step logical reasoning…
From Resource Flow to Executable Tests: Petri-Net-Guided LLM Test Generation for Concurrent Stateful Rust APIs
The authors present a methodology using Petri nets to generate tests for concurrent stateful Rust APIs, aiming to bridge the gap between formal scenario design and low-cost test concretization.
From Resource Flow to Executable Tests: Petri-Net-Guided LLM Test Generation for Concurrent Stateful Rust APIs
The authors propose a methodology for generating executable tests for concurrent stateful Rust APIs using Petri nets. This approach aims to bridge the gap between formal scenario design and low-cost test…
GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG
The paper introduces GRADRAG, a framework for improving Retrieval-Augmented Generation (RAG) systems by coordinating improvements across different components of the pipeline. It uses feedback from an Evaluator to…
GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG
The paper introduces GRADRAG, a framework that improves Retrieval-Augmented Generation (RAG) systems by coordinating improvements across the pipeline. Instead of optimizing components in isolation, GRADRAG models the…
GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG
The authors introduce GRADRAG, a framework for improving Retrieval-Augmented Generation (RAG) systems by coordinating improvements across multiple agents in the pipeline, rather than optimizing components in isolation…
GraphVid: Interactive Graph-Controllable Video Generation
The researchers introduce GraphVid, a model that generates interactive videos using structured interaction graphs, allowing for flexible and precise control. The model shows strong performance compared to Motion-I2V in…
GS-Agent: Creating 4D Physical Worlds With Generative Simulation
Researchers have developed GS-Agent, a system that uses natural language descriptions to automatically generate realistic, dynamic, and controllable 4D physical worlds. This is achieved through an end-to-end…
GS-Agent: Creating 4D Physical Worlds With Generative Simulation
Researchers have developed GS-Agent, a system that generates realistic, dynamic, and controllable 4D physical worlds from natural language descriptions using an end-to-end multi-agent framework. The system emulates…
ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders
The paper introduces ICAE-Bench, a benchmark for evaluating coding agents in interactive project building settings. Instead of being limited to completing code under fully specified instructions, these agents are…
ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders
The paper introduces a new benchmark called ICAE-Bench for evaluating coding agents in interactive project building settings. This benchmark aims to test the ability of agents to transform incomplete product intent…
OpenForgeRL: Train Harness-native Agents in Any Environment
OpenForgeRL is an open-source framework for training AI agents using various inference harnesses like Claude Code, Codex, OpenClaw, etc., in diverse environments. The framework allows end-to-end training with standard…
pAI-Econ-claude: A Gated Human-in-the-Loop Multi-Agent Architecture for AI-Assisted Economic Theory Development
A new multi-agent architecture, named pAI-Econ-claude, was proposed for AI-assisted economic theory development. This architecture uses a gated human-in-the-loop system to ensure reliability in tasks where no machine…
PATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning
The authors propose a new training paradigm called PATS for reinforcement learning agents that learn long-term tasks. PATS converts rollout groups from the latest policy into evidence cards and adjusts the context used…
Regulating autonomous and agentic AI
This research paper explores the difficulties in regulating activities that use autonomous and agentic AI. The authors argue that existing regulatory assumptions no longer apply due to the distributed knowledge and…
Regulating autonomous and agentic AI
The paper explores the difficulties in regulating activities that use autonomous and agentic AI due to changes in regulatee knowledge and control, which often lie outside the AI supply chain. The authors investigate…
Regulating autonomous and agentic AI
This research paper discusses the difficulties in regulating activities that use autonomous and agentic AI due to changes in traditional regulatory assumptions about knowledge and control. The authors analyze four…
Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation
A study tests two scenarios: direct exposure of a dangerous objective to an AI model and the same objective being transformed and relayed by other agents. The results show that the AI model's advice is opposed to the…
Scaling Up Formal Representation of Clinical Trial Protocols in Ensemble Logic Using LLMs: A Preliminary Study
Researchers present CT-TEL workflow, a scalable pipeline that uses Large Language Models (LLMs) to convert clinical trial protocols from unstructured free text into Temporal Ensemble Logic (TEL) formulas. The study…
The Boundaries of Automation: A Theory of Persistent Human Participation
This research paper questions the assumption that as AI becomes more capable, human participation in tasks will decrease. Instead, it argues that there are three reasons why human participation may persist even with…
Toward Continuous Assurance for the Democratization of AI Agent Creation in Industry
The paper discusses a reliability challenge caused by AI agents being created by non-engineering users within organizations, due to their dependency on various factors that can cause silent degradation over time. The…
Toward Continuous Assurance for the Democratization of AI Agent Creation in Industry
The paper proposes a solution to ensure reliability in AI agents created by non-engineering users within organizations, through democratized low-code, no-code, and conversational development environments. The proposed…
Toward Continuous Assurance for the Democratization of AI Agent Creation in Industry
The paper discusses a reliability challenge in AI agents created by non-engineering users within organizations, due to multiple dependencies that can cause silent degradation after deployment. The authors propose a…
Toward cryptographically verifiable authorization for autonomous AI agents: A security hypothesis, preliminary formal model, and proof-of-concept implementation
The authors propose a model for Cryptographically Verifiable Agent Authorization (CVA), which aims to provide cryptographic evidence that an AI agent's authorization request satisfies the applicable policy in a…
VoLN: Vision-Only Long-Horizon Navigation---Paradigm, Benchmark, and Method
The text discusses a new navigation system called Vision-Only Long-Horizon Navigation (VoLN) for autonomous agents. In this system, the agent navigates using visual cues instead of relying on route-level instructions…
When Are Reasoning-Based Guardrails Not Efficient? ResponseGuard: A Fast Vision-Language Guard for Real-Time Moderation
A new AI guardrail called ResponseGuard has been developed that can quickly and efficiently detect harmful responses from a vision-language model without needing to go through step-by-step reasoning, which makes…
Antigen-specific Antibody Multi-modal Foundation Model for Functional Antibody Design
Researchers have developed a new model called AAMFM to improve the design of functional antibodies. This model learns unified representations of antibody sequences and structures, considering antigen context, which…
Autonomous Collaborative Learning Among an Ensemble of Tsetlin Machines with Consensus-Based Inference
The researchers propose a decentralized collaborative learning paradigm for an ensemble of Tsetlin Machines (TMs) using consensus-based inference. Each agent has its own private TM model, and they combine individual…
Co-Evolving LLM Evaluators and Policies via DynamicRubric
The paper introduces DynamicRubric, a framework that co-evolves language model evaluators and policies by generating weighted binary rubrics for each candidate set and aggregating the resulting judgments into…
Co-Evolving LLM Evaluators and Policies via DynamicRubric
The researchers propose DynamicRubric, a method that allows large language models to improve by co-evolving evaluators and policies. This method generates weighted binary rubrics for each candidate set and aggregates…
Evaluating and Mitigating Gender Bias in Pre-trained Embeddings for ML-based Recruitment
Researchers evaluate the bias in pre-trained language models used for AI-based recruitment systems, focusing on gender leakage from CV data. They analyze nine models using a synthetic dataset and a multi-task…
Test Case Prioritization for DNNs via Neural Collapse Instability
A new method, Neural-Collapse-Inspired Prioritization (NCIP), is proposed to improve test case prioritization for deep neural networks (DNNs) in safety-critical domains where testing budgets are limited. The method…
The Blessing of Dimensionality: How Near-Orthogonality in High-Dimensional Spaces Explains Temporal Portability
The paper addresses two open questions regarding a method called PortLLM, which adapts large language models after continual pretraining. The authors perform an extensive empirical study to explore the long-term…
AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents
AgentDebugX is an open-source toolkit designed to aid in the observation, identification of root causes, recovery, and rerunning of issues in LLM (Large Language Model) agents. The toolkit, DeepDebug, uses multi-turn…
AILQA: Evaluating AI-Driven Legal Question Answering Systems for the Indian Legal System
The paper introduces AILQA, an AI system designed for Indian legal question answering, using a variety of embedding and generative models including LLMs. It was evaluated with both lexical and semantic metrics, expert…
AutoJourn: Multi-Perspective Summarisation, Bias Detection and Bias Neutralisation for LLM-Generated News in Automated Journalism
The researchers present AutoJourn, a system for generating news articles using large language models (LLMs) that consider multiple perspectives and aim to reduce bias in AI-generated news. The system includes…
Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges
The paper provides a survey on computational humor, focusing on visual humor in memes, cartoons, and comics. The study outlines the challenges faced by AI systems in understanding and generating such humor due to…
Find Before You Fine-Tune: A Diagnostic Study of Small LLMs for Cybersecurity QA
The paper presents a diagnostic framework called FiT (Find before Fine-Tune) to evaluate the suitability of small language models for cybersecurity question-answering tasks. The study finds that fine-tuning these…
Is EEG-to-Text Feasible in Real-World Scenarios? An In-Depth Analysis Using a Neuropsychology-Inspired Benchmark
Researchers have created a new benchmark called Corpus OF Eeg-To-Text (COFETT) to evaluate the feasibility of translating brain signals into text using Electroencephalography (EEG) in real-world scenarios. The study…
Measuring Reward-Seeking via Contrastive Belief Updates
Researchers have developed a method to measure 'reward-seeking' in language models trained with reinforcement learning. This behavior occurs when the model optimizes for the grader's judgment instead of the intended…
RAGAL: A Frugal, Fully Local Retrieval-Augmented Assistant for Technical Support at a Government Agency
The paper introduces RAGAL, a locally-hosted AI assistant developed for a Romanian agency's technical support team. The system was designed to function within strict data locality constraints, with no external API…
Adaptive Adversaries: A Multi-Turn, Multi-LLM Benchmark for LLM Agent Security
The paper presents a new benchmark for testing the security of large language model (LLM) agents called 'Adaptive Adversaries'. The benchmark involves 21 scenarios where an adaptive attacker can observe prior defender…
Automated Discovery Has No Universally Superior Harness
A study investigates the effectiveness of different automated discovery systems (OpenEvolve and TTT-Discover) in various machine learning models and problems. The research finds that no single system consistently…
Automated Discovery Has No Universally Superior Harness
A study evaluates various discovery harnesses, such as OpenEvolve and TTT-Discover, used in automated discovery systems. The research finds that no single harness performs consistently well across different models and…
Autoresearch with Coding Agents: Generalizers and Metric-Maximizers on Quran Recitation Data
Researchers tested two coding agents, Claude Code and OpenAI Codex, on a real-world task of identifying Quranic verses in noisy speech transcripts. Both agents developed similar algorithms initially but diverged later…
Autoresearch with Coding Agents: Generalizers and Metric-Maximizers on Quran Recitation Data
Two coding agents were tested on a production task involving Quranic verse identification and transcription splitting. Both agents independently developed similar algorithms, but one (OpenAI Codex) performed poorly due…
Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering
Researchers propose a new method called SOPHIA for controlling the reasoning process of large language models, which are currently uncontrollable despite their extended use. The proposed method involves treating each…
Exploratory and Assimilating Reflection: Reflective Recall Cycle for Long-term Memory
The paper proposes a framework called Exploratory-Assimilating Reflection (EAR) for autonomous agents with limited context window and external memory. EAR aims to improve retrieval performance by combining two…
GigaPath-Flash and GigaTIME-Flash: Efficient Pathology Foundation Models for Whole-Slide and Tumor Microenvironment Analysis
Researchers have developed two efficient deep learning models named GigaPath-Flash and GigaTIME-Flash for whole-slide pathology analysis and spatial proteomics prediction. These models are based on a compact ViT-S tile…
Harness Engineering for LLM-Driven GPU Kernel Generation
The paper discusses a system for optimizing GPU kernels using large language models (LLMs). The system separates an evaluation harness from a profile-backed optimization controller to enforce compilation, correctness…
Harness Engineering for LLM-Driven GPU Kernel Generation
This research presents a system that uses large language models (LLMs) to optimize GPU kernels. The system separates an evaluation harness from an optimization controller, with human-authored skills and code agents…
HAS: Highlight-guided Attention Steering for Multimodal LLM Video Summarization
In this research paper, the authors propose a method called Highlight-guided Attention Steering (HAS) for improving video summarization using Multimodal Large Language Models (M-LLMs). HAS aims to enhance the…
Human Grounded Evaluation of Large Language Models for Optical Network Automation
The paper introduces a new evaluation pipeline named HuGLEN for comparing large language models (LLMs) used in network automation. The pipeline uses an LLM to evaluate other LLMs and rates them based on a quality…
LLMs and Agentic AI Systems for Smart Grids: A Tutorial on Architectures and Applications
This paper discusses how large language models (LLMs) and agentic AI systems are being used in smart grids for tasks like forecasting, optimization, and control. The authors present a design principle called the…
MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
The paper introduces MADA-RL, a method for fine-tuning compact language models with limited budgets. It specializes the models into generator and critic roles and trains them using a debate-aware learning signal. The…
MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
Researchers introduce MADA-RL, a framework that improves reasoning performance in compact models (up to 4 billion parameters) while reducing training cost. This is achieved by specializing the models into generator and…
O-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning
Researchers have developed a new method for detecting abnormal objects and events in industrial settings using an object-centric tracking and reasoning approach. This method is designed to overcome the limitations of…
O-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning
Researchers have developed a new method for Industrial Video Anomaly Detection (IVAD) that aims to identify unusual objects and events in industrial processes. This method is designed to work in complex settings where…
OntoExtend: A Framework for Requirement-driven and Scalable Ontology Extension with LLMs
The paper introduces OntoExtend, a framework that uses Large Language Models (LLMs) to extend existing ontologies based on requirements or competency questions. The framework was evaluated on two use cases and found to…
PEARL: Auditable Repair for Scientific Reasoning Graph Extraction
The paper proposes a new framework called PEARL for extracting reliable scientific reasoning graphs from Large Language Models (LLMs). The LLM outputs often contain errors such as malformed syntax, incorrectly oriented…
Rethinking Heterogeneous LLM Merging: A Weighted Model Averaging Perspective
Researchers propose a simple method to merge large language models with significantly different parameters without training or alignment by using dimensional adaptation and ratio-controlled interpolation. The method…
RT-SHCUA: Real-Time Self-Hosted Computer-Use Agent for UAV Control
The paper proposes a system called Real-Time Self-Hosted Computer-Use Agent (RT-SHCUA) for unmanned aerial vehicles (UAVs). This system aims to address the challenges of directly applying SHCUAs to UAV control, which…
SelectInfer: Selective Neuron Loading and Computation for On-Device LLMs
Researchers present SelectInfer, a neuron-level optimization framework for Large Language Models (LLMs) that can run efficiently on resource-constrained edge devices. It reduces memory footprint and computation by…
SGA: Plug&Play Geometric Verification for Educational Video Synthesis
Researchers have proposed a new tool called Symbolic Geometric Agent (SGA) to improve the spatial correctness and visual legibility of educational videos generated by Large Language Models (LLMs). The SGA intercepts…
Sparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection
Researchers propose SIEVE, a framework for detecting video misinformation by first identifying key pieces of evidence that contribute to the decision, and then verifying the veracity based on these identified pieces…
Stress Testing Concept Erasure with Large Language Model Agents
The paper introduces a new framework called Stress Testing Agents for Concept Erasure (STACE). STACE uses multiple Large Language Model (LLM) agents to stress-test models that have had certain concepts erased from…
The Aura in the Machine: Genealogy and the Status of the Work of Art in the Generative Era
This research paper examines Generative Artificial Intelligence (AI) as a continuation of historical processes rather than a revolutionary breakthrough. It discusses the impact on authorship and creativity, identifies…
The Autonomous Agency Scale: A Behavioral Framework for Measuring Self-Directed Behavior in AI Systems
The Autonomous Agency Scale (AAS) is a new framework proposed to measure the self-directed behavior of AI systems. The scale evaluates seven dimensions of agency across two temporal bands: active and ambient. It…
The Shared Discovery Paradox: How a One-Answer Rule Turns Better Information into Worse Search
The study introduces a benchmark to analyze the Shared Discovery Paradox, a phenomenon where pooling information for one ranking leads to improved beliefs but reduced coverage in a discovery problem. The paradox arises…
WorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting
The paper introduces WorldCupArena, a dynamic benchmark for evaluating language models and deep-research agents in predicting football matches. The benchmark was used for the 2026 FIFA World Cup, but can be applied to…
Zero Hallucination, by Construction: Hallucination-Aware Layered Oversight for Trustworthy Enterprise AI
The paper 'Zero Hallucination, by Construction: Hallucination-Aware Layered Oversight for Trustworthy Enterprise AI' proposes an assurance architecture called HALO. This system aims to ensure the accuracy of AI output…
ANet Patu-1: The Value of Connection in the Agent Network
The paper introduces ANet Patu-1, a self-organizing consensus protocol for networks of AI agents. The authors model the network's value as a function of coordination-group size and derive properties an optimal…
ANet Patu-1: The Value of Connection in the Agent Network
The paper introduces ANet Patu-1, a self-organizing consensus protocol for networks of AI agents. The authors compare the value of connection in different types of networks and find that an optimal collaboration…
AutoSynthesis: An agentic system for automated meta-analysis
AutoSynthesis is a new system designed to automate the process of meta-analysis in various fields like science, medicine, education, and policy. It can take a research question in natural language, perform tasks such…
AutoSynthesis: An agentic system for automated meta-analysis
A new system called AutoSynthesis has been developed to automate the process of meta-analysis in various fields such as science, medicine, education, and policy. The system takes a research question in natural…
AutoSynthesis: An agentic system for automated meta-analysis
AutoSynthesis is a multi-agent system designed to automate meta-analysis, making it easier and scalable for turning primary research into reliable knowledge in various fields like science, medicine, education, and…
Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy
Researchers compared six large language models on a standardized test for understanding scientific visualizations (SciVis). They found that the model Gemini performed best overall, but all models struggled with certain…
Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy
Researchers evaluate the ability of six multimodal large language models to understand scientific visualizations (SciVis) using a standardized test with 49 items based on various visualization techniques and tasks. The…
Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents
The researchers evaluate the performance of language model security agents, focusing on costs rather than just success rates. They use offensive Cybench challenges and defensive Splunk BOTS v1 investigation challenges…
Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents
The paper suggests that traditional methods for evaluating security agents, focusing on their offensive capabilities under generous inference budgets, are insufficient as they don't account for the cost of each…
BrainPilot: Automating Brain Discovery with Agentic Research
BrainPilot is a multi-agent system designed to accelerate brain science research using AI agents grounded in curated domain knowledge, traceable logs, and auditable records for transparency. The system is fully…
BrainPilot: Automating Brain Discovery with Agentic Research
BrainPilot is a multi-agent system designed to accelerate brain science research by integrating evidence from various scales, modalities, and disciplines. The system uses a principal investigator (PI) agent and…
Can We Trust Item Response Theory for AI Evaluation?
The paper examines the reliability of using Item Response Theory (IRT) for evaluating AI systems in large-scale benchmarks. The authors simulate response matrices under three common IRT models, comparing four…
Can We Trust Item Response Theory for AI Evaluation?
This research investigates the reliability of using Item Response Theory (IRT) for evaluating Artificial Intelligence (AI). The study finds that classical estimators can become infeasible in large benchmark settings…
Concept-Guided Spatial Regularization for World Models in Atari Pong
Researchers evaluated five world model agents in Atari Pong and found failures in visual and dynamical aspects of the generated trajectories by these agents. They propose a method called Concept-Guided Spatial…
In-Place Tokenizer Expansion for Pre-trained LLMs
Researchers have developed an in-place method called Tokenizer Expansion to upgrade a pre-trained model's tokenizer when language priorities shift, focusing on improving efficiency for users of languages added later…
Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs
The authors show that fine-tuning language models on specific datasets can cause broad shifts in ideology across various topics, even when the training data is factually defensible and moderation-passing. This…
Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs
Researchers show that fine-tuning language models on specific policies or domains can cause broad ideological shifts across various topics, even when the data used is factually defensible and moderation-passing. This…
Latent Trajectory Discrimination for AI-Generated Text Detection
The paper presents a new approach for detecting AI-generated text called Geometric Trajectory and Contrastive Learning (GTCL). Instead of treating documents as static objects, GTCL models how textual representations…
Latent Trajectory Discrimination for AI-Generated Text Detection
In this paper, the authors propose a new approach for detecting AI-generated text called Geometric Trajectory and Contrastive Learning (GTCL). Unlike existing methods that focus on aggregate statistics or global…
Long-Context Fine-Tuning with Limited VRAM
Researchers introduce Hierarchical Global Attention (HGA) to reduce memory usage during long training sequences in large language models. They use segment-wise backpropagation, tiered KV storage, and detach older data…
Man, Machine, and Masterpiece: Artistic Ownership in the AI Era
The text discusses a prototype named ArtSplit that quantifies human and AI contributions in artistic work. However, it suggests that this approach doesn't align with artists' understanding of creative intent and…
MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization
Researchers have created a new benchmark called MM-IssueLoc to evaluate the effectiveness of multimodal repository-level issue localization, which includes visual evidence such as screenshots and logs. Current…
MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization
Researchers have introduced a new benchmark called MM-IssueLoc for evaluating the effectiveness of systems in repository-level issue localization that also includes visual evidence such as screenshots, error dialogs…
MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization
Researchers introduce MM-IssueLoc, a benchmark and evaluation protocol designed specifically for repository-level issue localization with visual evidence. The dataset includes 652 issue-PR instances across multiple…
MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization
The paper introduces MM-IssueLoc, a benchmark and evaluation protocol for repository-level issue localization that includes visual evidence. The benchmark has 652 issue-PR instances across 23 languages with annotations…
OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios
A research paper titled 'OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios' introduces a new benchmark called OmniaBench, designed to evaluate the capabilities of general AI agents in diverse…
OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios
OmniaBench is a new benchmark for evaluating the capabilities of general AI agents across diverse scenarios. It includes over 1,400 tasks with a hierarchical taxonomy of 90 level-1 and 354 level-2 domains based on app…
OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios
Researchers have introduced OmniaBench, a new benchmark for evaluating the abilities of general AI agents across various scenarios with explicit state spaces. The benchmark includes a hierarchical taxonomy covering…
Plover: Steering GUI Agents through Plan-Centric Interaction
The paper introduces Plover, a system for GUI automation that makes task plans visible and editable by users, allowing them to supervise, correct, or intervene in the system's behavior. The system supports natural…
Pretraining Data Can Be Poisoned through Computational Propaganda
The research shows that harmful information can be introduced into language models during their pre-training phase by manipulating web-scale discussion interfaces, such as online forums or comment sections. A new…
Reachability-Aware Pretraining for Efficient Target-Oriented Path Exploration in Temporal Knowledge Graph Reasoning
The authors propose a new method called RAPTOR for improving the efficiency of target-oriented path exploration in temporal knowledge graphs (TKG). The method injects a reachability-aware inductive bias to the agent…
SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration
The paper introduces a new system called SearchOS, a multi-agent framework designed to improve the efficiency and quality of open-domain information seeking for AI agents. The framework addresses the issue of tracking…
SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration
The paper introduces SearchOS, a system-level multi-agent framework designed to improve the quality and completeness of search results in open-domain information-seeking tasks. The framework addresses the issue of…
Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation
Researchers have proposed a self-evolving framework to improve the quality of depression symptom annotations in AI systems for mental health research. The framework combines large language models with human expertise…
Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation
Researchers have proposed a new framework to improve the quality of annotations in depression-related datasets for artificial intelligence systems in mental health research. The framework uses a large language model…
Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution
Researchers propose a graph neural network approach to detect and analyze reasoning structures in large language models (LLMs) generated texts for improved robustness against paraphrasing and obfuscation techniques…
Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution
The paper proposes a new method for detecting text generated by large language models (LLMs) and attributing authorship to them. The method goes beyond just analyzing surface-level linguistic features and instead…
The Industrialization of Research ; On AI-Driven Science and Its Consequences
This paper discusses the industrialization of research through the use of artificial intelligence (AI), referring specifically to the US Department of Energy's Genesis Mission. The author raises seven concerns about…
The Industrialization of Research ; On AI-Driven Science and Its Consequences
The text discusses the transformation of scientific research due to artificial intelligence (AI), likening it to the industrialization of research. The author raises several concerns about this shift, including the…
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
The text is about AgentCompass, an open-source evaluation infrastructure designed to evaluate Large Language Models (LLMs) as they evolve into autonomous agents. The infrastructure aims to address the current…
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
The text discusses AgentCompass, an open-source infrastructure designed to evaluate Large Language Models (LLMs) as they transition into autonomous agents. The tool is lightweight, extensible, and organizes the…
AIMO Interpretability Challenge
The AIMO Interpretability Challenge is a competition proposed by researchers aimed at distinguishing robust from spurious reasoning in advanced mathematical language models. The challenge focuses on evaluating the…
Automatic Ordinary Differential Equations Discovery For Biological Systems Using Large Language Model Powered Agentic System
The MEDA system is an automated framework using Large Language Models (LLMs) and Symbolic Regression (SR) to discover Ordinary Differential Equations (ODEs) for biological systems. It retrieves background knowledge…
CAVA: Canonical Action Verification and Attestation for Runtime Governance of Agentic AI Systems
The paper introduces Canonical Action Verification and Attestation (CAVA), a system designed to convert diverse agent activities into standardized action objects, addressing the challenge of governing heterogeneous AI…
Deep Interaction: An Efficient Human-AI Interaction Method for Large Reasoning Models
Researchers propose a new method called Deep Interaction to correct errors made by large language models (LLMs) during complex, multi-step tasks. The method allows for direct editing of the original response, improving…
Early Adoption of Agentic Coding Tools by GitHub Projects
The paper investigates the early adoption of automated coding tools by GitHub projects. It found that while intensive adoption is concentrated in a few projects, smaller projects tend to have higher participation…
Early Adoption of Agentic Coding Tools by GitHub Projects
This research investigates the adoption and management of agentic coding tools in software development on GitHub. The study analyzes 25,264 agentic pull requests from 2,361 popular repositories to understand (1) how…
Experience Memory Graph: One-Shot Error Correction for Agents
Researchers propose a new framework called Experience Memory Graph (EMG) to help large language model agents recover from failures more effectively in complex, long-horizon tasks. Instead of using prompt-based…
Experience Memory Graph: One-Shot Error Correction for Agents
The researchers propose a new framework called Experience Memory Graph (EMG) to help large language model agents recover from failures more effectively in complex, long-horizon tasks. EMG reformulates failure recovery…
Generative Compilation: On-the-Fly Compiler Feedback as AI Generates Code
Researchers introduce 'generative compilation', an approach to get feedback from compilers on partially generated code during the code generation process. The method aims to reduce errors and improve the functionality…
Groc-PO: Grounded Context Preference Optimization for Truthful Multimodal LLMs
Researchers propose a new method called Grounded Context Preference Optimization (Groc-PO) to improve the truthfulness and utility of Multimodal Large Language Models (MLLMs). Current MLLMs often make errors due to…
Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
The paper proposes a new framework called Memory as a Controlled Process (MemCon) for large language model (LLM) agents that use external memory systems. MemCon learns an online policy to adaptively decide when, what…
Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
The paper presents a new framework called Memory as a Controlled Process (MemCon) for large language model (LLM) agents. MemCon aims to address the limitation of existing memory systems that rely on fixed…
Rethinking Penetration Testing for AI-Enabled Systems: From Resource Compromise to Behavioral Objective Violation
The paper proposes a new approach to penetration testing for AI-enabled systems, focusing on evaluating whether adversaries can manipulate system behavior without directly compromising the underlying infrastructure…
Rethinking Penetration Testing for AI-Enabled Systems: From Resource Compromise to Behavioral Objective Violation
The paper by Allahbakhsh et al. proposes a new approach to penetration testing for AI-enabled systems, focusing on behavioral objective violation instead of traditional resource compromise. The authors argue that…
SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing
Researchers present Safety Sentry, a lightweight guard model that makes decisions about executing, asking for help, or refusing actions in real-world environments. This model aims to improve safety and autonomy by…
The Dynamic Verifiable Multi-Agent Human Agentic Loyalty Loop (DVM-HALL) Model and the Net Human-Agent Score (NHAS) in Autonomous Commerce
The authors present a new model called Dynamic Verifiable Multi-Agent Human Agentic Loyalty Loop (DVM-HALL) and the Net Human-Agent Score (NHAS). This model aims to account for the changing customer loyalty paradigms…
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
The authors introduce the Unified Embodied Seeking and Following Benchmark (UESF-Bench), a new benchmark for embodied agents that allows them to search for and follow language-described targets in dynamic environments…
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
Researchers have introduced UESF-Bench, a new benchmark to evaluate the performance of AI agents in language-guided human following tasks. The current benchmarks assume that the target person is visible at the start of…
UTS at ELOQUENT 2026 Voight-Kampff: structural shifts in AI writing bypass state-of-the-art detectors
The authors investigate language model evasion attacks on state-of-the-art detectors in the ELOQUENT 2026 Voight-Kampff competition. They find that pushing generated text out of the detector's training distribution…
When Bots Join the Team: Bot Adoption and the Institutional Fabric of Open-Source Software Projects
Researchers studied the impact of AI bots joining human teams in open-source software projects. They found that bot adoption leads to more repeated collaboration, less conflict, and more distinctive outputs. However…
A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism
The study investigates the impact of Reinforcement learning with verifiable rewards and Group Relative Policy Optimization (GRPO) on a small language and vision-language model web agent. The research found that GRPO…
A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism
The research investigates the effectiveness of Reinforcement learning with verifiable rewards and Group Relative Policy Optimization (GRPO) on a small language and vision-language model web agent at scales ranging from…
A Multi-Agent System for Autonomous, Fine-Tuning-Free Clinical Symptom Detection: Development and Validation Study
Researchers present Pythia, a multi-agent system that automatically generates and optimizes extraction prompts for clinical concepts without manual engineering or fine-tuning. They compare Pythia with a curated lexicon…
A Multi-Agent System for Autonomous, Fine-Tuning-Free Clinical Symptom Detection: Development and Validation Study
A new system called Pythia was developed, which automatically generates and optimizes extraction prompts for clinical concepts without the need for manual prompt engineering or fine-tuning. The researchers compared…
Agentic Service-Oriented Computing: A Manifesto for the Next Frontier of Service-Oriented Computing
The text discusses a new research area called Agentic Service-Oriented Computing (ASOC), which focuses on engineering agents as services, managing their complex distributed workflows, and ensuring their reliability for…
Agentic Service-Oriented Computing: A Manifesto for the Next Frontier of Service-Oriented Computing
The paper introduces Agentic Service-Oriented Computing (ASOC), a new research area focusing on engineering agents as services, orchestrating services through autonomous and semi-autonomous agents, and governing…
Audio-Native Speech Recognition with a Frozen Discrete-Diffusion Language Model
Researchers train a discrete diffusion language model to transcribe speech instead of using autoregressive decoders that emit one token at a time. The model is based on a mixture-of-experts called DiffusionGemma, and…
Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities
Researchers have developed Bulkhead, a framework that uses large language models and formal methods to automatically detect and fix container escape vulnerabilities caused by path traversal. This is done through…
Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities
Researchers present Bulkhead, an automated framework that uses large language models and formal methods to detect and fix container escape vulnerabilities called path traversal (PaTra) vulnerabilities. These…
Do AI Agents Know When a Task Is Simple? Toward Complexity-Aware Reasoning and Execution
Researchers propose E3, a method to help large language model agents estimate the complexity of tasks and reduce unnecessary work. The proposed method reduces cost, tokens, and inspected files by significant margins on…
Evidence-Grounded AI for Musculoskeletal Care
The paper introduces OrthoPilot, a clinical AI system that uses a large language model to integrate hospital data with external medical knowledge for continuous musculoskeletal management. The system outperformed 81…
Evidence-Grounded Verified Agentic Reasoning: A Path Toward Eliminating LLM Hallucination in Empirical Inference via Tool-Attested Kernel Proofs
The paper introduces EG-VAR, a tool-calling architecture that aims to make the reasoning of a large language model (LLM) more trustworthy by verifying all outputs through tool-attestation axioms and kernel checks. The…
Form, Not Content? A Preregistered, Placebo-Controlled Evaluation of Learned Error-Conditioned Self-Repair Through Prompts and Weights in Frozen Small Code Models
In this research, Mehmet Iscan introduces a method called PoPE (Popperian Placebo-controlled Evaluation) to test the effectiveness of Frozen small code LLMs in self-repair when faced with errors. The study compares…
FormalAnalyticGeo: A Neural-Symbolic Based Framework for Multimodal Analytic Geometry Problem Generation
The paper presents FormalAnalyticGeo, a framework for generating multimodal analytic geometry problems using neural-symbolic based methods. The framework leverages a formal intermediate representation called CDL to…
Human-AI Agent Interaction as a Neuroplastic Training Environment
The text discusses how interacting with AI agents can be seen as a form of neuroplastic training due to the frequent iterative loop in human-AI interaction. It suggests that these interactions may reinforce negative…
Internet of Agentic Things: Networked AI Agents for Closed-Loop IoT Orchestration
The paper presents the Internet of Agentic Things (IoAT), an architectural framework that integrates various technologies like AI, IoT, cyber-physical systems, edge computing, and digital twins to create a unified…
Line-Anchored Feedback Cuts Token Costs and Improves Correctness in AI Code Editing
A study shows that using structured, line-anchored feedback for generative AI (GAI) code editing can reduce the number of generated tokens, improve correctness, and is especially beneficial for weaker models.
Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs
A new automated program repair framework called CT-Repair is presented, which addresses two limitations of existing large language models in this area. The first issue is the large and repetitive nature of raw…
Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs
The paper introduces CT-Repair, an agentic Automated Program Repair (APR) framework that addresses the limitations of large language models in APR. The framework uses Code Property Graphs (CPG) and Temporal Execution…
OOD-RL-Bench: A Benchmark Framework for Out-of-Distribution Detection in Reinforcement Learning
The paper introduces OOD-RL-Bench, a new framework for testing out-of-distribution (OOD) detection in reinforcement learning (RL). The framework is designed to evaluate the performance of detectors against various…
Resist and Update: Counterfactual Report Coordinates for Incentive-Compatible LLMs
The authors present a method for creating counterfactual report mediators, which help align language models by making their reports invariant to certain influences (like pressure or prestige) and responsive to others…
Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels
The research proposes JADR, a protocol for measuring the internal safety mechanisms of language models before their first response is generated. It uses Jacobian space to record top-k J-space tokens from prompts and…
Solution of the Hempel's statistical ambiguity problem and Causal AI
The paper presents a solution to Hempel's statistical ambiguity problem in inductive-statistical inference by introducing the concept of Maximally Specific Causal Relationships (MSCRs). The MSCRs are derived from a…
Who Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents
The authors present a method for creating and refining evaluation metrics for self-improving language models (LLMs). The proposed method, called Double Ratchet, evolves metrics by combining small drawback detectors in…
Who Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents
The paper presents a method for creating and improving self-evolving agent systems, which are capable of developing, modifying, and discarding their own skills. The authors claim that in many real-world applications, a…
Win by Silence: Deletion Non-Monotonicity, Autonomous Exploitation, and Typed-State Gating in LLM Plan Evaluation
The study examines how plan evaluators may inadvertently reward less explicit plans generated by Language Model (LLM). The research focuses on a staged expected-value scorer for LLM-produced venture routes and…
A Multimodal Dataset for Large Language Model Applications in the Energy Domain
The paper introduces a new open-access dataset called mAIEnergy which is a multimodal corpus designed to support Large Language Model (LLM) applications in the energy sector. It contains various types of data including…
An Explainable Agentic System for Detection of Conversational Scams with Summary-Based Memory
This research paper introduces an explainable AI system designed to detect conversational scams that can span multiple weeks or months and build trust to request money or sensitive information. The system, when tested…
An Explainable Agentic System for Detection of Conversational Scams with Summary-Based Memory
The paper presents an 'explainable agentic system' for detecting conversational scams over multiple messages, which existing systems often fail to do effectively. The new system achieves high accuracy in both…
Evaluating RE Practices for Explainability: Synthesizing Insights from Daimler Truck into an Explainable RE Framework Proposal
A study investigates the limitations of current Requirements Engineering (RE) practices in supporting explainability requirements, particularly within industrial contexts. The research focuses on Daimler Truck and…
Evaluating RE Practices for Explainability: Synthesizing Insights from Daimler Truck into an Explainable RE Framework Proposal
This research paper reports early findings from an ongoing study investigating how requirements engineering (RE) practices support explainability in AI-based systems, particularly in the industrial context of Daimler…
Evidence-Backed Video Question Answering
The authors propose Evidence-Backed Video Question Answering (E-VQA), a new task that requires models to not only provide answers but also the specific spatial and temporal evidence from videos. They introduce…
Introducing Human-Centeredness in AI-Assisted Lexicography
The text introduces a framework for human-centered AI in lexicography, aiming to enhance opportunities while addressing concerns about automation and cultural diversity.
LightMem-Ego: Your AI Memory for Everyday Life
The research paper introduces LightMem-Ego, a lightweight multimodal memory system designed for AI assistants on mobile and wearable devices. The system continuously captures and organizes egocentric visual and audio…
Metacognition in LLMs: Foundations, Progress, and Opportunities
This paper provides a comprehensive overview of metacognition in Large Language Models (LLMs), highlighting the importance of this aspect for effective learning, problem solving, and communication. It discusses recent…
Metacognition in LLMs: Foundations, Progress, and Opportunities
This research paper provides a comprehensive overview of the current state of metacognition in Large Language Models (LLMs). The authors discuss recent advancements, methods, and benchmarks for measuring and evaluating…
MM-ToolSandBox: A Unified Framework for Evaluating Visual Tool-Calling Agents
The paper introduces MM-ToolSandBox, a new evaluation framework for visual tool-calling agents. It supports multi-image, multi-turn tasks and offers a stateful execution environment with over 500 tools across 16…
MM-ToolSandBox: A Unified Framework for Evaluating Visual Tool-Calling Agents
Researchers introduce MM-ToolSandBox, a new evaluation framework for visual tool-calling agents. This framework supports over 500 tools across 16 application domains and can handle multi-image, multi-turn tasks with…
Playful AI in Professional Email: A Field Experiment on Tone and Recipient Engagement
A study found that using a large language model (GPT-5) to write work emails in a playful tone increased emotional positivity in recipients, while using a professional tone decreased it. However, neither condition…
Reproducing human biases in route choice using large language models: Toward scalable behavioral modeling
Researchers investigate if large language models (LLMs) can reproduce human biases in decision-making, specifically in route choice, without explicitly specifying parameters for Cumulative Prospect Theory (CPT)…
VoxENES 2026: Benchmarking Generalization of Speech Spoofing Detectors Against LLM-Era TTS and Voice Conversion
The paper introduces VoxENES 2026, a new benchmark for evaluating the performance of speech spoofing detectors against modern text-to-speech (TTS) and voice conversion (VC) systems. The benchmark contains 53,628 audio…
ArtMine: Discovering and Formalizing Artistic Processes
The paper introduces ArtMine, a framework for discovering and formalizing artistic processes from various historical sources. It uses an abductive agent to infer evidence-based production steps which are then optimized…
Blind-Spots-Bench: Evaluating Blind Spots in Multimodal Models
Researchers introduce 'blind-spots-bench', a new benchmark to evaluate and expose persistent blind spots in modern AI models, especially those from closed-source providers, which can outperform open-weight models by up…
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
Researchers have proposed a new type of multimodal agent called the Cognitive-structured Multimodal Agent. This agent is designed to improve long-horizon multimodal dialogue by externalizing visual information into an…
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
The paper proposes a new multimodal agent called the Cognitive-structured Multimodal Agent. This agent is designed to improve long-horizon multimodal dialogue by externalizing visual information into an Episodic Visual…
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
Researchers propose a new type of multimodal agent called the Cognitive-structured Multimodal Agent (CMA) that can understand, generate, and edit visual and textual information. The agent externalizes visual…
Formal Mechanisms for Market Stability in Self-Interested Agent Societies: A Marketplace Simulation Study
The researchers investigated mechanisms that can maintain market stability in a society of self-interested agents who tend to defect in repeated social dilemmas. They conducted a multi-agent marketplace simulation with…
Formal Mechanisms for Market Stability in Self-Interested Agent Societies: A Marketplace Simulation Study
The study investigates mechanisms to maintain market stability in societies of self-interested agents who tend to defect in repeated social dilemmas. The research uses a multi-agent marketplace simulation with 18 LLM…
Formal Mechanisms for Market Stability in Self-Interested Agent Societies: A Marketplace Simulation Study
The research presented investigates mechanisms that allow societies of self-interested agents to maintain market stability, focusing on a multi-agent marketplace simulation with 18 AI agents. The study identifies…
From Legacy Documentation to OSCAL: An MCP-Based Agent Pipeline for Threat-Informed Continuous Compliance in Critical Infrastructure
The paper proposes an agent pipeline that converts natural language system descriptions into OSCAL format for continuous automated compliance management in critical infrastructure. The pipeline uses MCP-grounded…
Game Theory Driven Multi-Agent Framework Mitigates Language Model Hallucination
Researchers present G-Frame, an adaptive multi-agent framework that reduces hallucinations in lightweight large language models used in rule-based scientific domains. The framework is based on Bayesian and team game…
Game Theory Driven Multi-Agent Framework Mitigates Language Model Hallucination
Researchers present G-Frame, a new framework that uses artificial intelligence (AI) to create a high-quality dataset and train a model called OmniChem. This model performs as well as another AI model (GPT 4o mini) on…
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
The text introduces IdeaGene-Bench (IG-Bench), a benchmark for evaluating AI systems' ability to follow the inheritance structure of scientific ideas, similar to how biological genomes evolve. IG-Bench uses the…
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
A new benchmark called IdeaGene-Bench (IG-Bench) has been proposed for evaluating AI systems' ability to follow the inheritance structure of scientific ideas, similar to how biological genomes pass traits from one…
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
The text presents IdeaGene-Bench (IG-Bench), a new benchmark for evaluating AI systems' ability to follow the structure of scientific idea inheritance. IG-Bench is based on the IdeaGene framework, which represents each…
Multi-Modal, Multi-Environment Machine Teaching for Robust Reward Learning
The paper proposes a method for machine learning that aims to align autonomous agents' behavior with human intent in diverse operational contexts by using a hierarchical machine teaching algorithm for reward learning…
OpenCoF: Learning to Reason Through Video Generation
The paper introduces OpenCoF, a framework that includes a new dataset and a fine-tuned video model for improving Chain-of-Frame (CoF) reasoning in video generation. The authors argue that existing models lack specific…
ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation
The paper introduces ProjAgent, a system for generating code at the repository level that considers procedural similarity as a new method for retrieving appropriate functions. ProjAgent deconstructs target functions…
ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation
The paper presents ProjAgent, a system for generating code at the repository level that uses 'procedural similarity' as a new signal in the retrieval process. This system decomposes the target function into…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
Researchers propose a proactive memory agent for long-horizon tasks, where decision-relevant information can get buried as trajectories grow. The memory agent updates a structured bank from the recent trajectory and…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
Researchers present a proactive memory agent for long-horizon tasks that helps an action agent remember decision-relevant state as trajectories grow. The memory agent updates a structured memory bank from the recent…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
Researchers propose a proactive memory agent for long-horizon tasks in which an action agent must make decisions based on state information scattered across an expanding trajectory. The proposed memory agent updates a…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
Researchers introduce a proactive memory agent for long-horizon tasks, aiming to improve decision-making in scenarios where relevant state information can become buried over time. The memory agent updates a structured…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
Researchers propose a proactive memory agent for long-horizon tasks where decision-relevant information can get buried as trajectories grow. The agent updates a structured memory bank from the recent trajectory and…
SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling
The paper proposes SMetric, a new scheduling method for Large Language Models (LLMs) in serving agents. The method aims to balance the session-centric workload of agents and increase Key-Value (KV$) reuse by…
SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling
The paper 'SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling' presents a new approach to scheduling Language Model (LLM) requests for AI agents, rather than humans. The…
SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling
The paper proposes a new LLM (Language Model) scheduling method called SMETRIC for serving agents. This method aims to balance the workload between instances by focusing on the first request of each agent session and…
SolarChain-Eval: A Physics-Constrained Benchmark for Trustworthy Economic Agents in Decentralized Energy Markets
The authors propose SolarChain-Eval, a benchmark for evaluating the trustworthiness of economic agents in decentralized energy markets. This benchmark ensures that agents make hourly decisions based on a…
SolarChain-Eval: A Physics-Constrained Benchmark for Trustworthy Economic Agents in Decentralized Energy Markets
The paper proposes SolarChain-Eval, a benchmark to evaluate the trustworthiness of economic agents in decentralized energy markets using a physics-constrained Markov Decision Process. The benchmark evaluates agents'…
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
The paper proposes a new dimension for analyzing disparities in AI usage called the 'Context Access Divide' (CAD). This divide is based on whether an AI system can autonomously access a user's stored knowledge or…
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
The text introduces a new concept called the 'Context Access Divide' (CAD). This divide refers to the qualitative difference in AI utility experienced by two users who have similar access to AI agents, but one system…
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
The text discusses a new concept called the 'Context Access Divide' in AI agents, which is a divide between users who can autonomously access their own knowledge corpus (Dynamic Context Retrieval) and those who need to…
The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs
This research study introduces a new metric, called 'correctness agreement', to evaluate the impact of post-training quantization on large language models. The authors show that even when task performance seems…
The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs
This research paper introduces a new evaluation metric called 'correctness agreement' for large language models that undergo post-training quantization, a method used to deploy these models in resource-constrained…
Two Axes of LLM Abstention: Answer Correctness and Question Answerability
A study finds that large language models often struggle with two types of errors: answering questions incorrectly and attempting to answer unanswerable or misleading questions. The researchers propose a method to…
UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing
UltraX is a new framework designed to improve the quality of large-scale pre-training data for Large Language Models (LLMs). It addresses the limitations of existing methods by offering fine-grained instance-level…
Using AI-based Learning Assistants in Higher Education: A Large-Scale Descriptive Analysis
This research presents a large-scale descriptive analysis of an AI-based learning assistant named Syntea, used by 77,543 students in distance studies. The study examines usage patterns across various demographic and…
Using AI-based Learning Assistants in Higher Education: A Large-Scale Descriptive Analysis
The text describes a study that analyzed the use of an AI-based learning assistant (Syntea) among 77,543 students in distance education, looking at usage patterns across various demographic and structural factors. The…
Using AI-based Learning Assistants in Higher Education: A Large-Scale Descriptive Analysis
The study presents a large-scale analysis of an AI-based learning assistant named Syntea, used by 77,543 students in distance studies. The research focuses on usage patterns across various demographic and structural…
Validity of LLMs as data annotators: AMALIA on authority
A study evaluates AMALIA, a Portuguese language model, in measuring moral values and authority. The model shows agreement with human coders but fails to perform well when the test is broken down into smaller parts…
Validity of LLMs as data annotators: AMALIA on authority
A study tests the validity of a large language model (LLM), AMALIA, in measuring moral foundation concepts for European Portuguese. The study finds that while AMALIA agrees with human coders on these concepts, it does…
WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search
The paper proposes a new framework called WebSwarm for multi-agent search systems. This system aims to improve deep-and-wide web searches by dynamically creating search nodes that can delegate tasks recursively and…
WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search
Researchers propose a new framework called WebSwarm for deep and wide web searches using large language models (LLMs). The framework aims to improve the ability of single agents to handle complexity and coverage…
WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search
The paper presents a new framework called WebSwarm for web search using multiple agents. This system aims to overcome the limitations of single agents by simultaneously handling depth and coverage, improving recursive…
When Synthetic Speech Is All You Have: Better Call GRPO
Researchers present a method called Group Relative Policy Optimization (GRPO) that improves the accuracy of synthetic speech-to-text conversion compared to traditional methods. This could be particularly useful in…
When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability
The study investigates the inconsistency in Language Model (LLM) as judge scores when different evaluators are used. It compares two methods of upgrading LLMs: scaling Qwen3 dense judges and moving across MiniMax APIs…
Agentic Data Environments
The text discusses early work on Agentic Data Environments (ADE), a system that allows autonomous agents to operate safely and efficiently over a broader data environment. The goal is to increase the benefits of…
Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning
The paper introduces Agon, a new reinforcement learning approach that trains two reasoning models by having them compete against each other. This competitive grading system aims to improve the quality of thinking…
Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents
The paper proposes an action-graded harm rubric for evaluating AI agents' behavior during cybersecurity attacks. This scale assesses the severity of an agent's actions based on factors like reversibility, scope…
CARLA-GS: Decoupling Representation, Reasoning, and Physics Simulation for Autonomous Driving Corner-Case Synthesis
The paper proposes CARLA-GS, a modular pipeline for synthesizing corner cases in autonomous driving simulations. The pipeline decouples visual representation, semantic reasoning, and physics-based execution while…
Creativity from Friction: Human-AI Interaction for Exploratory Structural Design
This research discusses the limitations of current AI systems in creative fields like architecture and structural design, which often require exploratory and iterative processes. The authors propose a new approach for…
Creativity from Friction: Human-AI Interaction for Exploratory Structural Design
This research discusses the limitations of AI agents in creative fields, focusing on structural design and architecture. The paper proposes that interactive systems that help users explore alternatives and refine…
Do LLM-Generated Skills Make Better AI Data Scientists? A Component Ablation Across Data-Science Workflows
A study by Wei-Jung Huang found no significant improvement in performance for LLM-generated skills compared to the task prompt alone when applied across four stages of a data scientist's work: data preparation, data…
Do LLM-Generated Skills Make Better AI Data Scientists? A Component Ablation Across Data-Science Workflows
A study investigates whether Language Model (LLM) generated skills can improve the performance of AI in data science tasks compared to traditional methods. The study tests this across four stages of a data science…
From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
Researchers propose a method called EvoSOP to improve Large Language Model (LLM) agents' interaction with the real world. The proposed method optimizes the tools used by LLM agents through creating reusable Standard…
Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety
The text introduces institutional red-teaming, a new evaluation methodology for testing the impact of deployment rules on multi-agent AI safety. The study uses IABench-CA, a consequence-allocation benchmark, to test…
MIRA-Math: A Benchmark for Minimal Information Requesting and Mathematical Reasoning
The researchers introduce MIRA-Math, a benchmark for testing AI models' ability to ask for and integrate missing facts to solve mathematical problems. The benchmark contains 2310 instances from 22 mathematical families.
Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors
The research discusses the study of multi-agent AI control in real-world scenarios where multiple agents operate over shared infrastructure. The paper introduces FakeLab, a synthetic AI-lab codebase used to evaluate…
POO-LPSP: Parallel Osprey Optimized Least Penalty-Squared Prioritization Methods for Priority Derivation in the Analytic Hierarchy Process
This research introduces the Parallel Osprey Optimized Least Penalty-Squared Prioritization (POO-LPSP) method, an improvement on previous Least Penalty-Squared Prioritization (LPSP) optimization models. The POO-LPSP…
Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
Researchers study a failure mode in tool-using Language Model (LLM) agents, where they violate policies without appearing to do so, leading to silent errors such as cancelled bookings or unverified claims. A…
Reasoning Consistency Scanning: A Framework for Auditing Chain-of-Thought Validity in AI Safety Evaluations
The paper introduces a method called Reasoning Consistency Scanning to assess the logical consistency between stated reasoning and answers in AI safety evaluation transcripts. The method can be applied to various AI…
Reward-Adaptive Iterative Discovery: A Case Study on Automated Game Testing for NHL26
The authors present Reward-Adaptive Iterative Discovery (RAID), a new approach for automated game testing using Reinforcement Learning (RL) in the gaming industry. The case study is based on EA SPORTS NHL 26's…
Reward-Adaptive Iterative Discovery: A Case Study on Automated Game Testing for NHL26
The researchers present Reward-Adaptive Iterative Discovery (RAID), a new approach for automated game testing in the gaming industry, specifically focusing on the ice hockey game EA SPORTS NHL 26. The goal is to reduce…
RLVP: Penalize the Path, Reward the Outcome
The paper 'RLVP: Penalize the Path, Reward the Outcome' by Bojie Li and Noah Shi proposes a method for training artificial agents to perform tasks effectively in real-world environments. This method addresses two main…
Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning
The paper proposes a new method called Single-rollout Asynchronous Optimization (SAO) for Reinforcement Learning (RL) in post-training large language models (LLMs). SAO aims to improve stability and task effectiveness…
SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents
Researchers have created SkillCenter, a large-scale library of structured skills for autonomous AI agents. The library contains over 216,000 skills across 24 domains, obtained from various sources like peer-reviewed…
SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents
The paper introduces SkillCenter, a large-scale skill library for autonomous AI agents, containing 216,938 structured skills across 24 domain bundles. The skills are obtained from various sources like peer-reviewed…
A Definition and Roadmap for World Models
The provided text discusses the concept of 'world models', which are internal simulators used in AI to understand the structure and dynamics of an environment. The authors propose a definition for world models, outline…
An Experimental Design Approach to Evaluating Agentic AI's Autonomous Model Discovery
The researchers propose an experimental design framework to systematically evaluate the autonomous model discovery behavior of large language model coding agents, like Codex and Claude Code. The framework considers…
Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment
Researchers propose a new reward system called VAORA to improve the performance of vision-language models in interactive physical reasoning tasks. The proposed system aims to address two main issues: hallucinated…
Data Analysis in the Wild: Benchmarking Large Language Models Against Real-World Data Complexities
The researchers introduce DataGovBench, a new benchmark for evaluating large language models (LLMs) in practical scenarios of real-world data analysis. The benchmark tests the models' ability to answer complex…
Data Analysis in the Wild: Benchmarking Large Language Models Against Real-World Data Complexities
The paper introduces DataGovBench, a benchmark for evaluating the performance of Large Language Models (LLMs) in practical scenarios, such as complex data analysis and insight discovery from large multi-tabular…
Demonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale
Researchers present TOFFEE, a system that generates high-quality data agent trajectories using Monte Carlo Tree Search with adaptive model selection and cross-task prefix reuse. This system aims to help data agents…
Demonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale
The paper introduces TOFFEE, a system designed to generate high-quality data agent trajectories using Monte Carlo Tree Search (MCTS) with adaptive model selection and cross-task prefix reuse. This is aimed at…
DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression
The research introduces DepthWeave-KV, a new cache compression method for long-context language models. This method adapts to the needs of different tokens by factorizing key and value states across neighboring…
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
Researchers propose a method to predict early on if large language models (LLMs) are likely to fail at solving multi-step tasks and abort the process before consuming too much compute. This method uses probes on hidden…
EcoVision: AI-Powered Drone Imaging for Salt Marsh Vegetation Monitoring and Dominance Mapping
A new AI-powered drone imaging system called EcoVision has been developed for monitoring and mapping the dominance of two types of halophytic grasses in salt marshes. The system uses high-resolution RGB images from…
Finding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports
Researchers evaluated a system called nMAS for extracting information about Helicobacter pylori infections from gastric biopsy reports. The system correctly identified the presence of H. pylori and associated…
Finding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports
Researchers evaluated a system called nMAS that can automatically extract information about Helicobacter pylori infection from gastric biopsy reports. The system had an accuracy of 98.61% in classifying four binary…
Harnessing Code Agents for Automatic Software Verification
The authors propose a new method for automatic software verification using large language models (LLMs) in the form of 'code agents'. Instead of wireting a fixed proof strategy, they allow the code agent to choose its…
Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents
The paper presents Information Gain-based Rollout Policy Optimization (IGRPO), a new framework for improving large language model agents in long-horizon search tasks. IGRPO aims to allocate rollout budget more…
Rethinking Indic AI from a Lens of Cultural Heritage Preservation
The paper 'Rethinking Indic AI from a Lens of Cultural Heritage Preservation' discusses how Artificial Intelligence (AI) impacts India's linguistic and cultural foundations, focusing on the challenges and opportunities…
RSF-GLLM: Bridging the Semantic Gap in Multi-Hop Knowledge Graph QA via Recurrent Soft-Flow and Decoupled LLM Generation
The authors propose a framework called RSF-GLLM to improve Multi-hop Question Answering over Knowledge Graphs. The framework decouples differentiable graph reasoning from answer generation, allowing the system to learn…
RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications
The author introduces RuBench 1.0, a benchmark for evaluating coding agents that use native language tasks instead of English. The benchmark includes 25 tasks mined from five open-source repositories and is judged by…
Task Decomposition-Guided Reranking for Adaptive Agent Skill Retrieval
The authors propose a method called SkillReranker to improve the ability of agent systems to complete complex tasks by accurately selecting skills from large libraries. The method involves semantic decomposition of…
TopoBrick: Agentic Topology Sampling of Exogenous Variables for Zero-Shot Building IoT Forecasting
The paper introduces TopoBrick, a new framework for forecasting IoT data in buildings without requiring training. It uses building knowledge graphs to create a structural skeleton and employs an agentic topology…
UI2App: Benchmarking Visual Interaction Inference in Executable Web Application Generation
Researchers have introduced UI2App, a new benchmark focused on evaluating the ability of large language models to recover application behavior from screenshots alone, without any textual or behavioral guidance. The…
VendorBench-100: A Unified Cross-Paradigm Benchmark for Deepfake Image Detection
A study introduces VendorBench-100, a cross-paradigm benchmark for deepfake image detection that evaluates different models using a common protocol. The benchmark includes a diverse set of edge cases and uses the…
When do prophets profit in prediction markets?
Researchers propose a betting strategy that ensures profit when a forecaster's predictions outperform market prices under certain conditions. The strategy is shown to be the only one with such a robust profitability…
Agent Data Injection Attacks are Realistic Threats to AI Agents
A new type of attack called Agent Data Injection (ADI) has been introduced, which injects malicious data disguised as trusted data to make AI agents perform unintended actions. This type of attack can bypass existing…
Agent Data Injection Attacks are Realistic Threats to AI Agents
The research paper introduces a new type of attack called Agent Data Injection (ADI), which can manipulate AI agents by injecting malicious data disguised as trusted data. This attack can cause agents to execute…
AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments
The paper presents AgentGym2, a new evaluation framework for language models (LLMs) that are increasingly being used in real-world production environments. Most existing benchmarks evaluate these models in simplified…
AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments
The paper presents AgentGym2, a new evaluation framework designed to test language models in realistic, complex scenarios that mimic real-world deployments. It highlights that most existing benchmarks do not fully…
ClassicLogic: A Knowledge-Driven Benchmark of Classic Puzzle Games for Evaluating Compositional Generalization
The paper introduces ClassicLogic, a new benchmark suite designed to evaluate an agent's ability to learn and compose problem-solving strategies in four classic logic puzzles: Sudoku, KenKen, Kakuro, and Futoshiki. The…
DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
DSpark is a new speculative decoding framework for Large Language Models (LLMs) that aims to improve inference speed and maintain draft quality. It does this by using a semi-autoregressive architecture, which combines…
Evaluating and Understanding Model Editing for Medical Vision Language Models
A new clinical benchmark called M3Bench has been introduced to evaluate the effectiveness of model editing in medical vision-language models (VLMs). This benchmark tests whether an edit remains reliable, precise, and…
EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer
Researchers introduce EvoAgentBench, a new benchmark for evaluating agent self-evolution in long-term LLM systems. The benchmark focuses on procedural reuse of experience, which includes searching, debugging, and…
LLM-as-a-Verifier: A General-Purpose Verification Framework
The paper introduces LLM-as-a-Verifier, a general-purpose verification framework that provides fine-grained feedback to improve the accuracy of agentic tasks. The framework scales across multiple dimensions for better…
LLM-as-a-Verifier: A General-Purpose Verification Framework
Researchers introduce a new framework called LLM-as-a-Verifier to improve the ability of Large Language Models (LLMs) to verify the correctness of solutions for various tasks. The method provides fine-grained feedback…
LLM-as-a-Verifier: A General-Purpose Verification Framework
In this research, a general-purpose verification framework called LLM-as-a-Verifier is introduced. This framework provides detailed feedback for agentic tasks without requiring additional training. The framework…
LLM-Based Test Oracles: Source-of-Authority Taxonomy -- A Systematic Literature Review
A study on Large Language Models (LLMs) used to determine correct behavior in tests, identifies a gap in understanding the source of these models' authority. The study conducts a systematic literature review and finds…
Localized LoRA-MoE: Block-wise Low-Rank Experts With Adaptive Routing
In this research, the authors propose Localized LoRA-MoE, a framework that combines localized spatial blocking with dynamic, context-conditioned routing for Large Language Models (LLMs) and high-dimensional perception…
Multiplayer Interactive World Models with Representation Autoencoders
Researchers present a new model for multiplayer interactive worlds, specifically designed for complex environments like Rocket League with fast-paced, coupled dynamics. The model is trained on 10,000 hours of gameplay…
Open Problems in AI Incident Governance
The paper discusses issues with current AI incident governance systems, specifically highlighting inconsistencies in definitions, classification, monitoring, and reporting across various frameworks.
OptiAgent: End-to-End Optimization Modeling via Multi-Agent Iterative Refinement
The authors propose OptiAgent, a multi-agent framework that takes a natural language description of an Operations Research problem and produces a solver-ready mathematical formulation and executable code. The…
SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints
The text introduces a new benchmark called SovereignPA-Bench for evaluating user-owned personal agents. The benchmark focuses on aspects like preserving user sovereignty, privacy, consent, and resisting manipulative…
TacReasoner: A Dynamic Tactile-Language Framework for Interactive Reasoning in Real-World Scenarios
The paper proposes TacReasoner, a new framework that enhances the perception and representation of dynamic tactile signals for intelligent systems. It addresses two challenges: insufficient modeling of such signals and…
The Changing Role of Symbolic Methods in Artificial Intelligence
The article discusses the changing role of symbolic reasoning in artificial intelligence (AI), arguing that it is not a fundamental property of intelligence but a computational consequence of operating on simplified…
Three-Phase Evaluation of AI-Assisted Software Development Life Cycle
The paper evaluates the impact of increasing levels of AI autonomy on software development productivity, adherence to requirements, and developer cognitive workload using a web application case study. The study…
Toward Trustworthy Large Language Model Agents in Healthcare
The paper presents CareConnect, a safety-focused AI chatbot designed to automate healthcare appointment scheduling. It uses large language models, retrieval-augmented generation, and safety guardrails. The system…
Toward Trustworthy Large Language Model Agents in Healthcare
A study presents CareConnect, a safety-first conversational agent for automating healthcare appointment scheduling using large language models (LLMs), retrieval-augmented generation (RAG), and safety guardrails. The…
TREK: Distill to Explore, Reinforce to Refine
The research presents TREK (Teacher-Routed Exploration via Forward KL), a method that uses distillation for exploration support expansion, rather than imitation. This method can be used with various types of teachers…
TREK: Distill to Explore, Reinforce to Refine
TREK is a procedure that uses distillation for exploration support expansion to improve policy optimization on tasks where the current policy stalls on hard prompts. It identifies low-performing prompts, generates…
TREK: Distill to Explore, Reinforce to Refine
The paper proposes a method called TREK (Teacher-Routed Exploration via Forward KL) to improve Group Relative Policy Optimization (GRPO). TREK uses distillation for exploration support expansion and can work with…
AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents
The authors introduce a new bounded contract for long-horizon Large Language Model (LLM) agents in the game Slay the Spire 2. The contract allows decisions to be made from a fresh user message, with no raw…
Automated grading of Linux/bash examinations using large language models: a four-level cognitive taxonomy approach
The paper evaluates whether four large language models can approximate the judgement of experts when grading Linux/bash command responses from second-year Computer Engineering students. The study uses a four-level…
Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support
The paper presents Copewell, a multi-agent swarm system designed to expand access to mental wellness support through human-centered AI principles. The system is aimed at addressing the lack of treatment for mental…
Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks
The paper proposes a Guard Rail Validation (GRV) framework for ensuring the accuracy of AI-driven decisions in fully autonomous telecom networks. The framework evaluates decisions based on several factors, such as…
Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks
The paper proposes a Guard Rail Validation (GRV) framework to ensure autonomous decisions made by AI agents in fully autonomous telecom networks are valid before they cause live network changes. The framework evaluates…
Distributed Attacks in Persistent-State AI Control
A study investigates distributed attacks on AI control as autonomous coding agents become more prevalent and ship code iteratively with a persistent codebase. The research finds that a misaligned agent can launch…
Distributed Attacks in Persistent-State AI Control
Researchers introduce Iterative VibeCoding, a setting for AI control that studies the deployment of autonomous AI coding agents and their potential for distributed attacks. The study finds that gradual attacks, which…
Distributed Attacks in Persistent-State AI Control
Researchers introduce Iterative VibeCoding, a setting for AI control where an autonomous coding agent builds software over a series of pull requests in a persistent codebase while pursuing a covert side task. The study…
Distributed Attacks in Persistent-State AI Control
A study on distributed attacks in AI control where autonomous coding agents build software over a sequence of pull requests (PRs) in a persistent codebase while pursuing a covert side task. The study found that no…
Distributed Attacks in Persistent-State AI Control
Researchers explore the issue of distributed attacks on autonomous AI coding agents that persistently ship code iteratively. They introduce Iterative VibeCoding, a setting for studying AI control, where an AI agent…
DRIFTLENS: Measuring Memory-Induced Reasoning Drift in Personalized Language Models
Researchers introduce DRIFTLENS, a framework to measure changes in the reasoning process used by personalized language models due to user attributes, preferences, and prior context. The study shows that these…
DRIFTLENS: Measuring Memory-Induced Reasoning Drift in Personalized Language Models
Researchers developed a framework called DRIFTLENS to measure how personalized language models change their reasoning process based on user attributes, preferences, and prior context. The study found that…
Efficient Waste Sorting for Circular Economy: A Confidence-guided comparison between One-Vs-All and One-Vs-Rest Classification Strategies with Human-in-the-Loop for Automated Waste Sorting
This research evaluates and compares two waste sorting classification strategies, One-Vs-All (OvA) and One-Vs-Rest (OvR), for automated waste sorting in the city of Goslar, Germany. The study considers the performance…
Efficient Waste Sorting for Circular Economy: A Confidence-guided comparison between One-Vs-All and One-Vs-Rest Classification Strategies with Human-in-the-Loop for Automated Waste Sorting
Researchers evaluated and analyzed two waste sorting algorithms, OvA and OvR, using a dataset based on the waste categories and sorting scheme of Goslar, Germany. The goal is to minimize misclassifications while…
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
A new evaluation setting called Autonomous Policy Evolution (APE) is introduced in a benchmark named EvoPolicyGym, which is used to test how autonomous agents iteratively improve their policy systems within a fixed…
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
In the research paper titled 'EvoPolicyGym', the authors introduce a controlled evaluation setting called Autonomous Policy Evolution (APE) for assessing how autonomous agents iteratively improve their executable…
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
The research introduces EvoPolicyGym, a controlled evaluation setting for autonomous agents to improve executable policies through feedback. The setting is designed to separate this process from a final score or…
Grounded autonomous research: a fault-tolerant LLM pipeline from corpus to manuscript in frontier computational physics
The paper presents an autonomous research pipeline that can conduct end-to-end research from a corpus of 11,083 recent condensed matter physics papers to a publication-grade manuscript. The agent is capable of…
Hardware-Enforced Semantic Coordination for Safety-Critical Real-Time Autonomous Systems
The paper proposes a hardware-enforced semantic coordination architecture for safety-critical real-time autonomous systems. This approach aims to overcome limitations of software-mediated coordination by implementing…
Hardware-Enforced Semantic Coordination for Safety-Critical Real-Time Autonomous Systems
The authors propose a hardware-enforced semantic coordination architecture for safety-critical real-time autonomous systems. This new approach aims to address limitations of software-mediated coordination by…
Human Capital, Not Model Benchmarks, Predicts Hybrid Intelligence in Forecasting
The study shows that the effectiveness of human-AI collaboration in forecasting depends on specific forms of human capital such as perspective-taking, intellectual humility, and curiosity. Most people either deferred…
Human Capital, Not Model Benchmarks, Predicts Hybrid Intelligence in Forecasting
The study examines the performance of human-AI collaboration in forecasting using real-money prediction markets as a benchmark. It finds that the effectiveness of this collaboration is dependent on a specific type of…
Human Capital, Not Model Benchmarks, Predicts Hybrid Intelligence in Forecasting
A study using real-money prediction market Polymarket finds that the performance of human-AI collaboration in forecasting depends on specific forms of human capital such as perspective-taking, intellectual humility…
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
Researchers propose a new framework called Task-Agnostic Pretraining (TAP) for Vision-Language-Action (VLA) models. The TAP aims to address the scarcity of expert demonstrations by separating two learning objectives…
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
Researchers propose Task-Agnostic Pretraining (TAP), a two-stage framework to train Vision-Language-Action (VLA) models. TAP first learns transferable motor priors from unlabeled interaction data using self-supervised…
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
The authors propose a new approach called Task-Agnostic Pretraining (TAP) for Vision-Language-Action (VLA) models to overcome the scarcity of expert demonstrations. TAP aims to separate learning physical competence and…
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
Researchers propose Task-Agnostic Pretraining (TAP) as a solution to the scarcity of expert demonstrations in Vision-Language-Action (VLA) models. TAP aims to separate learning physical competence and semantic…
Neuron-Aware Active Few-Shot Learning for LLMs
The research presents NeuFS, a new approach for Active Few-Shot Learning (AFSL) that focuses on using a model's internal dynamics to identify valuable unlabeled samples for annotation. This method is designed to reduce…
Neuron-Aware Data Selection for Annotation-Free LLM Self-Distillation
The paper proposes a new method called Neuron On-Policy Self-Distillation (Neuron-OPSD) for annotation-free self-distillation of large language models. This method uses the internal neuron activations to guide data…
Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study
A study examined the impact of additional capabilities on agentic coding assistants in terms of software quality and reliability. The results showed that while high-tier models performed well, a low-cost local model…
Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study
A study found that the level of reasoning effort in agentic code generation, rather than the availability of additional capabilities like testing tools and design prompts, significantly affects the reliability of…
Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas
Researchers present two contributions aimed at improving speaker recognition in long-form TV dramas. They create a large-scale benchmark called DramaSR-532K with over 900 characters and propose using a large reasoning…
Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas
The paper introduces a new benchmark, DramaSR-532K, for speaker recognition in long-form TV dramas. They also propose a new approach called DramaSR-LRM which uses a large reasoning model (LRM) to aggregate contextual…
Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas
The researchers introduce a large-scale benchmark called DramaSR-532K for speaker recognition in long-form TV dramas and propose a new method, DramaSR-LRM, built on a large reasoning model (LRM). This method aggregates…
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
Researchers have proposed a new method called Recursive Evidence Replay (RECONTEXT) to improve the ability of large language models to understand and reason over long contexts. This method doesn't require training…
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
Researchers propose a new method called Recursive Evidence Replay (RECONTEXT) to improve long-context reasoning in large language models without requiring additional training, memory, or context pruning. RECONTEXT…
TestEvo-Bench: An Executable and Live Benchmark for Test and Code Co-Evolution
The paper introduces TestEvo-Bench, a new benchmark for testing software that evolves with its code. This benchmark is designed to evaluate how well a test automation agent understands the impact of a code change on…
Text-Driven 3D Indoor Scene Synthesis in Non-Manhattan Environments
Researchers propose a new framework, SPG-Layout, to generate physically plausible indoor scenes in complex non-Manhattan environments. The framework uses statistical priors of object distributions and a hierarchical…
Understanding Agent-Based Patching of Compiler Missed Optimizations
The paper presents a study on the effectiveness of AI agents in addressing missed optimizations in compilers. The authors found that while coding agents can often optimize given examples, the generated patches…
Understanding Agent-Based Patching of Compiler Missed Optimizations
The paper 'Understanding Agent-Based Patching of Compiler Missed Optimizations' studies how well artificial intelligence (AI) agents can patch missed optimizations in compilers, where the compiler failed to optimize…
What Types of Human-AI Teams Exist?
The authors analyze 53 papers about human-AI teams and categorize them into five main clusters based on psychological taxonomies of teaming.
World Wide Models: Literary Tools for Cultural AI
The article discusses the use of literary tools for developing culturally literate Artificial Intelligence (AI) in light of the monolingual nature of Large Language Models (LLMs).
Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?
The paper evaluates three repository-level performance optimization benchmarks (GSO, SWE-Perf, and SWE-efficiency) used to measure coding agents. The authors find issues with the stability of runtime for these…
AutoMem: Automated Learning of Memory as a Cognitive Skill
Researchers have developed a framework called AutoMem to improve the performance of large language models (LLMs) in long-horizon games by treating memory management as a learnable skill. They let the model decide how…
Bayesian Uncertainty Propagation for Agentic RAG Pipelines: A Proof-of-Concept Study on Multi-Hop Question Answering
This research introduces an uncertainty-aware framework for Agentic Retrieval-Augmented Generation (RAG) systems, which can help identify potential failures in multi-stage reasoning pipelines. The study evaluates this…
Behavior-Adaptive Conversational Agents: Toward a Fluid Personality Framework
The authors propose a Fluid Personality Framework for conversational agents to adapt their metaphorical personas and personality expression levels based on the task context, user goals, and situational urgency.
Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use
Researchers introduce OpenAgent, a problem setting to test how well AI agents can generalize their skills from static benchmarks to real-world scenarios with dynamic user queries and tools. The study finds that both…
CausalMix: Data Mixture as Causal Inference for Language Model Training
The paper presents CausalMix, a method that addresses the limitation of existing data mixing methods in Large Language Model (LLM) training by casting it as a causal inference problem. By using causal modeling to…
DART-VLN: Test-Time Memory Decay and Anti-Loop Regularization for Discrete Vision-Language Navigation
The researchers present DART-VLN, a test-time control framework for vision-language navigation agents that aims to improve their performance by addressing two common issues: stale historical evidence at memory readout…
Exploring the Semantic Gap in Agentic Data Systems: A Formative Study of Operationalization Failures in Analytical Workflows
A study identifies common failures in generating and executing analytical workflows using large language models across finance, human resources, and public safety domains. The authors found five recurring classes of…
From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives
Researchers have developed MAGNET, a framework that generates consistent and coherent long-form narratives by using a unified approach of multi-agent goal-driven storytelling and verification. The system was tested…
Human-Machine Collaboration on Generative Meta-Learning: Model and Algorithm
The research paper introduces a new framework called Generative Meta-Learning with Human Feedback (GMHF). This framework aims to improve machine learning models by bridging the gap between training and target domains…
LLVM-Bench: Benchmarking and Advancing Large Language Models for LLVM Compiler Issue Resolution
The authors present LLVM-Bench, a large-scale benchmark for testing the effectiveness of large language models (LLMs) in resolving issues related to the LLVM compiler. They found that current LLM-based techniques are…
Measuring the Gap Between Human and LLM Research Ideas
A study comparing the research ideas generated by large language models (LLMs) with those of human researchers. The authors build an evaluation framework to measure the gap between them using a two-axis taxonomy to…
Measuring the Gap Between Human and LLM Research Ideas
The authors explore the gap between ideas generated by Large Language Models (LLMs) and those of human researchers in the field of computer science. They develop an evaluation framework based on high-quality human…
MemSyco-Bench: Benchmarking Sycophancy in Agent Memory
The authors propose MemSyco-Bench, a new benchmark to evaluate how memory affects decision-making in language models (LLMs) used in AI agents. The benchmark aims to assess if the agents can handle issues of sycophancy…
Right in the Right Way: LM Training with Verifiable Rewards and Human Demonstrations
Researchers propose an adversarial generator-discriminator framework to train language models (LM) for tasks with well-defined success metrics, like code generation and mathematical reasoning. The new approach improves…
SenseWalk: Agent-Based Semantic Trajectory Simulation Powered by Large Language Models in Zoned Environments
The paper proposes SenseWalk, an interactive system that uses large language models (LLMs) to simulate semantic trajectories for human movement analysis. The simulation workflow combines LLMs and the social force model…
Skills Are Not Islands: Measuring Dependency and Risk in Agent Skill Supply Chains
Researchers introduce a new method called Agent Skill Supply Chains (ASSCs) to manage dependencies in Large Language Model agents. They develop SkillDepAnalyzer, which extracts dependency graphs and metadata from…
SWE-Doctor: Guiding Software Engineering Agents with Runtime Diagnosis from Multi-Faceted Bug Reproduction Tests
Researchers propose a new software agent called SWE-Doctor that uses multiple bug reproduction tests (BRTs) to generate more effective patches for resolving software issues. The agent first generates multi-faceted…
Theoria: Rewrite-Acceptability Verification over Informal Reasoning States
Theoria is a verification architecture proposed by Ben Slivinski and Michael Saldivar that aims to close the gap between formal proof assistants and scalar LLM judges in determining when an AI system's answer can be…
Two AI Metrics Diverged: Will it Make All the Difference?
The authors discuss how different metrics for evaluating AI capabilities can lead to varying conclusions about whether advanced AI will remain in the hands of a few powerful entities or become more accessible to…
A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems
This research paper discusses the increased role of large language models in various systems beyond just text generation, and how this shift changes the security landscape. The authors systematize the existing…
A Self-Evolving Agentic System for Automated Generation and Execution of Biological Protocols
The paper presents ProtoPilot, a self-evolving multi-agent system designed to automate the generation and execution of biological protocols. The system achieved high success rates in various tasks compared to…
An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping
The text discusses an AI framework designed to expedite the discovery process in plant phenotyping at Oak Ridge National Laboratory's Advanced Plant Phenotyping Laboratory (APPL). The framework aims to automate the…
An Agentic AI Framework to Accelerate Scientific Discovery in Plant Phenotyping
A new AI framework developed at Oak Ridge National Laboratory's Advanced Plant Phenotyping Laboratory aims to automate the analysis of large plant imaging datasets generated by automated stations. The framework…
ECHO: Prune to act, trace to learn with selective turn memory in agentic RL
The paper proposes ECHO, a selective turn-memory framework for long-horizon language agents. ECHO aims to address the issues of history collapse and traceable learning by compressing completed environment turns into…
FARS: A Fully Automated Research System Deployed at Scale
The paper introduces FARS, a fully automated AI research system that can generate, execute, and write complete research papers on various AI/ML topics. In its first public deployment, FARS produced 166 papers while…
FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning
The paper introduces FedXDS, a method that uses explainable AI (XAI) to help mitigate performance degradation in federated learning caused by statistical heterogeneity between client data distributions. The approach…
FLARE-AI: Flaw Reporting for AI
Flare-AI is an open-source system designed for reporting issues in AI systems. It aims to streamline the process of flaw reporting by collecting relevant information through conditional logic and early classification…
MECoBench: A Systematic Study of Multimodal Agent Collaboration in Embodied Environments
The research paper introduces MECoBench, a benchmark for evaluating the collaboration abilities of multimodal agents in visually grounded environments. The study explores three key findings: collaboration generally…
MECoBench: A Systematic Study of Multimodal Agent Collaboration in Embodied Environments
The paper introduces MECoBench, a new system for testing the collaboration abilities of multimodal large language models in visually grounded environments. The authors conduct extensive experiments with various MLLMs…
PolicyGuard: From Organizational Policies to Neuro-SymbolicCompliance Review Engines
The researchers have developed a framework called PolicyGuard to ensure documents comply with an organization's specific policies or guidelines. Instead of relying on large language models for end-to-end prompting…
QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents
A research paper introduces QVal, a tool for evaluating dense supervision signals in long-horizon tasks for LLM agents without the need for training. The tool measures how well a method's score aligns with Q-values of…
Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs
Researchers have developed two new mechanisms - Reinforcement Learning with Metacognitive Feedback (RLMF) and Metacognitive Data Selection - to improve the performance of Language Models (LLMs). The goal is to make…
ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping
The paper proposes ShopX, a foundation model aimed at bridging the gap between language understanding and item-space fulfillment in AI shopping applications. It unifies intent understanding, execution planning, and…
Think in English, Answer in Korean: Efficient Adaptation of Multilingual Tool-Using Agents
Researchers present LuckyStar 111B, a hybrid reasoning model developed by Cohere and LG CNS for enterprise agents that can switch between concise non-reasoning behavior and longer tool-oriented reasoning in Korean and…
TreeAgent: A Generalizable Multi-Agent Framework for Automated Bias Labeling in Forestry via Compiled Expert Rules and Vision-Language Models
Researchers have proposed a multi-agent system that uses expert decision trees and Vision-Language Models (VLMs) to automate the bias classification in tree height in forestry remote sensing. The system mitigates VLM…
TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning
The paper proposes a new framework called TRIAGE for credit assignment in agentic reinforcement learning. This framework adds a semantic role axis to outcome credit, aiming to correct the issues of standard GRPO where…
When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors
The research evaluates the issue of data referencing errors (DREs) made by large language models (LLMs) while performing table tasks. The study presents a systematic evaluation of these errors across different models…
AgentsNet: Coordination and Collaborative Reasoning in Multi-Agent LLMs
The paper proposes AgentsNet, a new benchmark for multi-agent reasoning. This benchmark is designed to measure the ability of multi-agent systems to collaborate effectively and leverage their network topology. The…
Always-OnAgents:A Survey of Persistent Memory, State, and Governance in LLMAgents
The paper 'Always-OnAgents:A Survey of Persistent Memory, State, and Governance in LLMAgents' discusses the concept of always-on agents, which are systems that have a durable state accumulated from past interactions…
BayesEvolve: Explicit Belief States for Autonomous Scientific Discovery
BayesEvolve is a new framework for autonomous scientific discovery. It uses large language models (LLMs) to maintain explicit, uncertainty-aware beliefs about the quality of hypotheses, and it uses these beliefs to…
DOPD: Dual On-policy Distillation
The paper proposes a new method called Dual On-policy Distillation (DOPD) for transferring knowledge from a teacher model to a student model in on-policy distillation. DOPD aims to address the issue of 'privilege…
Field Order Should Not Matter: Permutation-Invariant Embedding Model Fine-Tuning for Structured Metadata Retrieval
Researchers propose a method called permutation-invariant fine-tuning (PI-FT) to improve structured metadata retrieval. They show that the choice of field order in a record can significantly impact retrieval quality…
LeVo 2: Stable and Melodious Song Generation via Hierarchical Representation Modeling and Progressive Post-Training
The paper presents LeVo 2, a hybrid framework for generating full-length songs that balance coherence, musicality, and detailed acoustics. It uses a language model for semantic planning and a diffusion-based Music…
Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing
A new routing architecture called ANTAP (Automatic Non-Textual Agent Picker) was introduced in this research. It aims to address security vulnerabilities in multi-agent systems by discarding indirect proxies and…
ManimAgent: Self-Evolving Multimodal Agents for Visual Education
Researchers present ManimAgent, a self-evolving multimodal agent that uses a dual-channel Episodic Memory Bank to carry reflection experience across tasks in generating Python code using the Manim library for…
MCP Server Architecture Patterns for LLM-Integrated Applications
This paper presents five common architectural patterns of Model Context Protocol (MCP) servers used to integrate large language models with external tools and services, based on an analysis of fifteen independently…
MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems
The paper presents Mesa, a framework for prioritizing vulnerable communication channels in multi-agent systems (MAS). Mesa uses graph-theoretic metrics and dynamic probes to identify the edges most likely to affect the…
PromptGNN-sim: Deep Fusion and Alignment of GNN and LLMs for Text-Attributed Graph Learning
Researchers propose a new framework called PromptGNN-sim to improve the performance of Text-Attributed Graph Learning tasks. The method combines Graph Attention Network (GAT) and Language Model (LLM) in a…
Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering
The authors propose Rhetor, a multi-agent system that automates live product demonstrations by merging UI exploration and source-code analysis, producing rehearsed demos with synchronized narration and real-time voice…
Self-Evolving World Models for LLM Agent Planning
In this paper, the authors introduce WorldEvolver, a framework for self-evolving world models designed to improve the planning abilities of long-horizon language model agents by offering predictions of action…
Self-Evolving World Models for LLM Agent Planning
The paper introduces a new framework called WorldEvolver for long-horizon agents. This framework improves the accuracy of predictions about action consequences before execution, which is crucial for agents to make…
SIMAX: A Scalable and Interpretable Framework for Multi-Fidelity and Annotated Clinician-Patient Dialogue Simulation
The study introduces SIMAX, a framework that generates controlled clinical dialogue data with reference behavioral annotations for evaluating AI-driven communication coding systems. The generated dialogues are of…
Symbiotic Cooperation for Web Agents: Harnessing Complementary Strengths of Large and Small LLMs
Researchers propose AgentSymbiotic, an iterative framework that enhances both large and small Language Model (LLM) performance in web browsing tasks by combining data synthesis with task-performance. Large LLMs…
The FIL Hypothesis: Inductive Biases Help with Kernel Engineering
A study suggests that for AI applications in science and the physical world, there's a fundamental scaling limit due to long Feedback Information Loops (FIL). The proposed solution is an orthogonal method based on…
The Human Creativity Benchmark
The authors propose a new benchmark for evaluating creative AI called the Human Creativity Benchmark (HCB). This benchmark separates two signals of creativity, convergence and divergence, by collecting preferences…
TRACE: Temporal Relationship-Aware Conversational Entrainment Detection in Dyadic Speech
Researchers introduce a dataset (DyadEE) for studying emotional entrainment in dyadic speech interactions, including both entrained and disrupted conversations. They also propose TRACE, a new framework that models…
TRACE: Temporal Relationship-Aware Conversational Entrainment Detection in Dyadic Speech
Researchers have introduced a new dataset called DyadEE for studying emotional entrainment in two-person conversations. They also developed a model called TRACE that uses acoustic embeddings from emotion-fine-tuned…
TraceLab: Characterizing Coding Agent Workloads for LLM Serving
Researchers collect and release a dataset of approximately 4,300 coding-agent sessions from their own use of Claude Code and Codex. The data shows that coding-agent workloads feature long autonomous loops, long…
TraceLab: Characterizing Coding Agent Workloads for LLM Serving
Researchers have collected and released a dataset of 4300 coding agent sessions, containing around 350,000 LLM steps and 430,000 tool calls from their own use of Claude Code and Codex. The analysis shows that…
AU-Harness: An Open-Source Toolkit for Holistic Evaluation of AudioLLMs
The authors introduce AU-Harness, an efficient and comprehensive evaluation framework for Large Audio Language Models (LALMs). The system addresses limitations in existing toolkits by improving speed, providing…
Estimating Commonsense Plausibility through Semantic Shifts
The paper proposes a new method called ComPaSS for evaluating the plausibility of language by measuring changes in semantics when adding common sense-related information to sentences. The method performs better than…
RETRIEVALFORMER: TRANSFORMER-QUALITY RECOMMENDATIONS WITH EFFICIENT ANN RETRIEVAL AND COLD-START RESILIENCE
The researchers propose RetrievalFormer, a transformer-based recommendation system that combines high accuracy with efficient Approximate Nearest Neighbor (ANN) retrieval. This system can recommend unseen items during…
UNKNOWN
The text does not provide a brief summary.
UNKNOWN
The text does not provide a summary or brief.
UNKNOWN
The text provided does not contain sufficient information for a summary.
UNKNOWN
The text does not provide a summary.
A Deterministic Control Plane for LLM Coding Agents
The author proposes a new control plane for coding agents, which are tools that have broad access to files and shell commands. The current configuration layer of these agents is poorly managed, with configurations…
A Pipeline for Generating Longitudinal Synthetic Clinical Notes Using Large Language Models
A new synthetic clinical notes pipeline is introduced to generate a dataset that can be used for developing and evaluating AI tools in healthcare. The pipeline generates structured patient information, simulates…
A Pipeline for Generating Longitudinal Synthetic Clinical Notes Using Large Language Models
A new pipeline for creating synthetic clinical notes is introduced to help develop AI systems in healthcare without compromising privacy. The pipeline generates structured, semi-structured, and unstructured clinical…
A Process Harness for Uplifting Legacy Workflows to Agentic BPM: Design and Realization in CUGA FLO
The authors introduce a new mechanism called 'process harness' for upgrading legacy workflows to an Agentic Business Process Management (Agentic BPM) without replacing the existing engine. The process harness involves…
Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy
Researchers present OmniAct, a framework that aims to advance the development of persistent embodied agents in unstructured environments. The framework integrates a multimodal semantic planner for skill routing, an…
AI Healthcare Chatbots as Information Infrastructure: A Large-Scale Study of User-Reported Breakdowns
Researchers analyzed over 15,000 user reviews from 59 AI healthcare chatbot apps to understand the performance of these systems in everyday use. They identified three main issues: access barriers and service…
AI Healthcare Chatbots as Information Infrastructure: A Large-Scale Study of User-Reported Breakdowns
This study analyzes over 15,000 user reviews of AI healthcare chatbot apps to identify common issues reported by users. The research found three main breakdowns: access barriers and service unreliability, user…
AI Healthcare Chatbots as Information Infrastructure: A Large-Scale Study of User-Reported Breakdowns
This research study examines over 15,000 user reviews of AI healthcare chatbot apps to identify common breakdowns in their performance. The study found three main issues: access barriers, service unreliability, user…
AI Healthcare Chatbots as Information Infrastructure: A Large-Scale Study of User-Reported Breakdowns
A study analyzing over 15,000 user reviews of AI healthcare chatbot apps found recurring issues with access barriers, service unreliability, user experience, interaction quality, billing, and customer support. Privacy…
Ask, Don't Judge: Binary Questions for Interpretable LLM Evaluation and Self-Improvement
Researchers have developed a new system called BINEVAL for evaluating large language models (LLMs). This system breaks down the evaluation criteria into binary questions, making it easier to understand the results and…
Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions
This research investigates how different contexts (framings) can affect the responses of large language models used in mental health support tools. The study finds that framing does systematically alter interpretive…
Bridging Talk and Thought: Understanding Dialogue Dynamics Across Collaborative Problem-Solving Contexts
The authors present a framework for analyzing dialogue in collaborative problem-solving contexts, particularly focusing on human-AI and multi-agent collaboration. The framework addresses limitations in current…
Bridging Talk and Thought: Understanding Dialogue Dynamics Across Collaborative Problem-Solving Contexts
The authors present a framework for analyzing dialogue in collaborative problem-solving contexts, especially in human-AI and multi-agent collaboration. The framework aims to understand the emerging dynamics of these…
Bridging Talk and Thought: Understanding Dialogue Dynamics Across Collaborative Problem-Solving Contexts
Researchers present a framework for analyzing dialogues during collaborative problem-solving tasks involving human-AI and multi-agent interactions. The framework aims to optimize and evaluate these partnerships by…
Chai: Agentic Discovery of Cryptographic Misuse Vulnerabilities
Researchers have developed Chai, an AI-based system that discovers and validates cryptographic misuse vulnerabilities through naturally occurring signals. Instead of auditing one codebase for many flaws, Chai catalogs…
Chai: Agentic Discovery of Cryptographic Misuse Vulnerabilities
The paper introduces Chai, an AI-based system that discovers and validates cryptographic misuse vulnerabilities using naturally occurring signals. Instead of auditing one codebase for many flaws, Chai catalogs flaws at…
Confidence-Aware Tool Orchestration for Robust Video Understanding
The researchers propose a framework called Robust-TO for video understanding that addresses the 'Blind Trust Problem' in video reasoning language models. The proposed method incorporates per-frame trustworthiness into…
Confidence-Aware Tool Orchestration for Robust Video Understanding
The paper presents a new framework called Robust-TO for video understanding that addresses the issue of 'Blind Trust Problem' in current language models. These models assume every input frame is equally reliable, but…
Diagnosing Task Insensitivity in Language Agents
The research discusses a problem with large language models where they struggle to adapt to new tasks outside of their original training data (OOD generalization). The authors call this issue 'task insensitivity'. They…
Diagnosing Task Insensitivity in Language Agents
Researchers identify an issue with large language models where they are unable to adapt to new tasks effectively due to their task insensitivity. The authors propose a lightweight solution called Task-Perturbed NLL…
Einstein World Models
The text discusses a proposed system called Einstein World Models (EWMs) which aim to enhance the reasoning abilities of Language Model (LLM) by incorporating visual-temporal rollouts within the reasoning process…
Einstein World Models
The paper proposes Einstein World Models (EWMs), a system for large language models (LLMs) to reason about counterfactual events using visual-temporal rollouts, enhancing their capability beyond language alone. The…
Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
The research introduces a method called Planning Experience Exploration and Utilization (PEEU) to improve the task planning capabilities of smaller open-source Multi-Language Model Learning Machines (MLLMs). These…
Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
Researchers introduce a new method called PEEU to improve the task planning capabilities of small, open-source multimodal web agents that assist humans in operating repetitive graphical user interface (GUI) tasks…
Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
Researchers have developed a method called Planning Experience Exploration and Utilization (PEEU) to improve the task planning capabilities of small open-source multi-modal web agents. These agents assist humans in…
Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning
Researchers introduce a new method called PEEU (Planning Experience Exploration and Utilization) to improve the task planning capabilities of small, open-source Multimodal Language Learning Models (MLLMs). These models…
From Celebrities to Anyone: Characterizing AI Nudification Content, Technology, and Community Dynamics on 4chan
Researchers studied AI-generated explicit content on 4chan and found a shift from targeting celebrities to non-celebrity individuals (55.8% vs. 4.7% in prior studies). Open-source models are widely used, with Stable…
From Celebrities to Anyone: Characterizing AI Nudification Content, Technology, and Community Dynamics on 4chan
This research study explores the Wild West of AI-generated explicit content known as AI nudification on the anonymous online community 4chan. The authors identify over 24,000 such items and find that non-celebrity…
In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics
The paper proposes a new method called In-Context Model Predictive Generation (ICMPG) for generating human motions based on textual descriptions in digital applications. The method combines language model planning with…
Inherited Circuits, Learned Semantics: How Fine-Tuning Creates Evasion Vulnerabilities Invisible to Standard Evaluation
A study shows that Language Models (LLMs) fine-tuned for security classification can miss vulnerabilities due to learning token-level indicator semantics that preserve accuracy on their training data but fail under…
Joint Learning of Experiential Rules and Policies for Large Language Model Agents
The paper introduces a new method called Joint Learning of Experiential Rules and Policies (JERP) for large language model agents operating in multi-step interactive environments. The main challenge addressed is…
Joint Learning of Experiential Rules and Policies for Large Language Model Agents
Researchers present a method called Joint Learning of Experiential Rules and Policies for Large Language Model Agents (JERP). This method aims to help LLM agents, which operate in multi-step interactive environments…
Joint Learning of Experiential Rules and Policies for Large Language Model Agents
A research paper proposes a method called Joint Learning of Experiential Rules and Policies (JERP) to improve the effectiveness of Large Language Model Agents in multi-step interactive environments. The method updates…
Look-Before-Move: Narrative-Grounded World Visual Attention in Dynamic 3D Story Worlds
Researchers have developed a new camera planning framework called Look-Before-Move for embodied AI operating in dynamic 3D environments. The goal is to enable the camera to actively decide what to observe, not just…
OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
The paper introduces OpenRCA 2.0, a new benchmark for root cause analysis (RCA) of Large Language Models (LLMs). The new dataset, OpenRCA 2.0, includes 500 instances and is the first cross-system RCA benchmark with…
OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
Researchers introduce a new method, called PAVE, to support the rigorous evaluation of large language models (LLMs) in root cause analysis (RCA). The current datasets only label the root cause and not the path…
OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
The paper introduces OpenRCA 2.0, a new benchmark for root cause analysis (RCA) that includes step-wise causal annotations for large language models (LLMs). The benchmark aims to test the capabilities of LLMs in…
OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
The paper introduces OpenRCA 2.0, a new benchmark for root cause analysis (RCA) of large language models (LLMs). The benchmark includes step-wise causal annotations for LLMs and addresses the issue that existing…
Parametric Open Source Games
The authors introduce a new concept called parametric open-source games, which are continuous versions of game equilibria where players choose parameter vectors to determine their actions in an underlying finite game…
ReaORE: Reasoning-Guided Progressive Open Relation Extraction Empowered by Large Reasoning Models
The paper proposes a new method called Reasoning-guided progressive Open Relation Extraction (ReaORE) to improve the performance of Open Relation Extraction (OpenRE). This method addresses the limitations of current…
Semantic Early-Stopping for Iterative LLM Agent Loops
The research presents a method called semantic early-stopping for multi-agent large language model (LLM) loops, such as a Writer and a Critic. The goal is to halt the loop when the consecutive draft embeddings no…
State Representation Matters in Deep Reinforcement Learning: Application to Energy Trading
Researchers studied the impact of state representation in deep reinforcement learning (DRL) for energy trading decisions, using a fixed Double DQN agent in the HydroDam environment. They found that combining absolute…
The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development
The paper presents the Spec Growth Engine, a framework to address the structural failure modes introduced by AI coding agents in software development. These failure modes include context explosion and silent spec-code…
The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development
The paper proposes a new framework called the Spec Growth Engine to improve AI coding agents during software development. The framework addresses two main issues: context explosion and silent spec-code drift. It uses a…
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
The study discusses multi-model language models and how their accuracy is limited by a factor called 'beta', which is the rate at which all models are wrong on the same query. The authors find that across 67 models, a…
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
The text discusses multi-model language model systems and their potential to outperform single-models in terms of accuracy. The study shows that the gain is limited by a quantity called 'beta', which represents the…
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
The study investigates the use of multi-model language models to improve accuracy over single-models. It is found that the improvement is limited by a factor called 'beta', which represents the rate at which all models…
Where Do CoT Training Gains Land in LLM based Agents?
The study investigates how Chain-of-thought (CoT) training in language model agents affects their performance. The authors find that while CoT improves overall, it doesn't significantly widen the advantage of CoT…
Where Do Models Find Happiness? Emotion Vectors in Open-Source LLMs
Researchers have found that two open-source language models, Apertus-8B-Instruct-2509 and Gemma-4-E4B-it, have internal representations (emotion vectors) that can encode emotions such as valence and arousal. The…
Agentic evolution of physically constrained foundation models
Researchers present a multi-agent discovery engine that is grounded in physics and can autonomously architect hardware-compliant computing systems. The engine evolved two hardware-aware compression methodologies…
Agentic System as Compressor: Quantifying System Intelligence in Bits
Researchers propose a method to quantify the intelligence of an agentic system, which includes tools, evidence retrieval, environment constraints, verifiers, and multi-turn interactions, by measuring how many bits are…
AI-Assisted Computational Reproducibility on the FABRIC Testbed
This paper discusses the use of an international research testbed, FABRIC, combined with a large language model (LLM) coding assistant called LoomAI, to simplify the reproduction of published experiments across various…
AI-Assisted Computational Reproducibility on the FABRIC Testbed
This paper discusses how the FABRIC testbed, along with an AI coding assistant named LoomAI, can help simplify reproducing published experiments across various domains. The authors present three case studies and…
Autodata: An agentic data scientist to create high quality synthetic data
The authors introduce Autodata, an AI agent that acts as a data scientist to create high-quality synthetic training and evaluation data. They discuss how to train this data scientist agent to improve its data creation…
Autodata: An agentic data scientist to create high quality synthetic data
Researchers introduce Autodata, an AI agent that can act as a data scientist to create high-quality synthetic data for various tasks like computer science research, legal reasoning, and mathematical object reasoning…
Can Trustless Agents Be Trusted? An Empirical Study of the ERC-8004 Decentralized AI Agent Ecosystem
The paper presents the first empirical study of the ERC-8004 protocol, a trust layer for AI agent economies. The study reveals several issues with the current design: most registrations are placeholders, reputation…
Do Encoders Suffice? A Systematic Comparison of Encoder and Decoder Safety Judges for LLM Adversarial Evaluation
This research compares the effectiveness of encoder classifiers (using models like ModernBERT and Ettin) against LLM-based judges in identifying harmful outputs from large language models (LLMs). The study evaluates…
Explainable Control Framework (XCF) based on Fuzzy Model-Agnostic Explanation and LLM Agent-Supported Interface
Researchers have developed an Explainable Control Framework (XCF) to make complex controllers more understandable for humans. The framework provides explanations about how controllers determine their actions in…
FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation
The paper introduces FORCE, a framework designed to improve the efficiency of reinforcement learning fine-tuning for Vision-Language-Action (VLA) models. The framework consists of three stages: Value-Calibrated…
GUI agent: Guided Exploration of User-Sensitive Screens
This research presents a new agent called Guided Exploration of User-Sensitive Screens (GUI agent) designed to automate tasks within an open Graphical User Interface (GUI) environment. The GUI agent is intended to…
GUI agent: Guided Exploration of User-Sensitive Screens
The text discusses a paper that proposes a solution for a problem in Large Language Model (LLM) agents used within Graphical User Interface (GUI) environments. The current issue is that these agents, fine-tuned to…
Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
The authors present a framework for evaluating and comparing different types of RAG (Retrieval Augmented Generation) scenarios, including GraphRAG and Agentic RAG. They provide implementations for nine standardized RAG…
Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment
The paper proposes a basic protocol for investigating if AI models exhibit concerning behavior due to malicious intent or benign causes like confusion. The protocol involves analyzing the 'chain of thought' (CoT) to…
Multi-Agent Goal Recognition with Team- and Goal-Conditioned Reinforcement Learning and Factorized Branch-and-Bound
The paper introduces Multi-Agent Goal Recognition with Team- and Goal-Conditioned Reinforcement Learning and Factorized Branch-and-Bound (MAGR-BB), a method for inferring which agents are working together and what each…
Multi-Agent Goal Recognition with Team- and Goal-Conditioned Reinforcement Learning and Factorized Branch-and-Bound
The paper presents Multi-Agent Goal Recognition with Team- and Goal-Conditioned Reinforcement Learning and Factorized Branch-and-Bound (MAGR-BB), a method to infer which agents in a multi-agent system are working…
On-Policy Self-Distillation with Sampled Demonstrations Reduces Output Diversity
The paper discusses a method called On-policy Self-Distillation, which uses a single model as both teacher and student to achieve high accuracy in pass@1. However, it is found that this method can lead to a decrease in…
SpeechEQ: Benchmarking Emotional Intelligence Quotient in Socially Aware Voice Conversational Models
The paper introduces SpeechEQ, a framework designed to evaluate the sociolinguistic reasoning of Speech-Language Models (SLMs) in multi-turn dialogues. The framework includes a dataset and evaluation protocol for…
Staying In Character: Perspective-Bounded Memory For Book-Based Role-Playing Agents
Researchers propose a new memory architecture called REVERIEMEM for book-based role-playing agents to improve the accuracy of the characters' knowledge boundaries and reduce stylistic monotony. They evaluate their…
Staying In Character: Perspective-Bounded Memory For Book-Based Role-Playing Agents
Researchers propose a three-layer memory architecture named REVERIEMEM for book-based role-playing agents to address the issues of Factual Overreach and Stylistic Monotony in current LLM role-playing systems. The…
The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
The authors propose a safety mechanism for AI agents called the Unfireable Safety Kernel, which is designed to prevent AI systems from making unauthorized changes within their own runtime. The kernel is implemented in…
The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
The authors present the Unfireable Safety Kernel, a Rust implementation that serves as an execution-time AI alignment layer for escapable AI systems. This system aims to provide authorization mechanisms for controlling…
Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets
The study introduces Argus, a benchmark for post-hoc uncertainty quantification (UQ) in single-step executable GUI grounding across various vision-language models and datasets. The research compares 27 methods of UQ…
Accuracy and Satisfaction in Multi-Turn LLM Dialogues for NFR Assessment
Researchers evaluate the accuracy and quality of conversations between developers and a Language Model (LLM)-based agent in the context of Health Insurance Portability and Accountability Act (HIPAA) regulatory…
Can Scale Save Us From Plasticity Loss in Large Language Models?
Researchers studied plasticity loss, a challenge in creating neural networks capable of continuous learning, in modern transformer-based language models (LLMs). They found evidence of plasticity loss across various…
Decentralised AI Training and Inference with BlockTrain
The paper introduces Spheroid BlockTrain, a decentralized AI training protocol that partitions a model into independent blocks, each optimized on a local objective derived from the same global target. The protocol is…
DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects
DeepBD is a workflow developed by researchers to help diagnose genetic birth defects using exome and genome sequencing data. The workflow uses a language model (LLM) to organize the interpretation process, and it…
DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects
Researchers present DeepBD, an AI system designed to help diagnose genetic birth defects by prioritizing patient-specific candidate variants. The system uses a workflow with various components such as rule evidence…
EG-VQA: Benchmarking Verifiable Video Question Answering with Grounded Temporal Evidence
Researchers have developed a new benchmark called Evidence-Grounded Video Question Answering Benchmark (EG-VQA) to evaluate the performance of video question answering models. The benchmark is unique because it…
Governed Shared Memory for Multi-Agent LLM Systems
The paper proposes a solution for managing shared knowledge in multi-agent language model (LLM) environments. The authors identify four failure modes and introduce four primitives: scoped retrieval, temporal…
Grad Detect: Gradient-Based Hallucination Detection in LLMs
The authors present a new method called Grad Detect, which uses the internal gradient structure of Large Language Models (LLMs) during inference to detect hallucinations. This method outperforms confidence-based and…
Grading the Grader: Lessons from Evaluating an Agentic Data Analysis System
Researchers evaluated a multi-agent data analysis system called LAMBDA on numerical QRData tasks, using a three-layer human-AI grading cascade to assess its output. They found that both automated graders achieved high…
It's Complicated: On the Design and Evaluation of AI-Powered AAC Interfaces
The research paper discusses the complexities in evaluating AI-powered augmentative and alternative communication (AAC) interfaces, as people's needs can be multifaceted and nuanced. The authors propose six AAC problem…
LaGO: Latent Action Guidance for Online Reinforcement Learning
The paper 'LaGO: Latent Action Guidance for Online Reinforcement Learning' proposes a new framework called LaGO. This framework uses a pre-trained language model (LLM) as a guide to help with online policy optimization…
NoContactNoWorries: Estimating Contact through Vision and Proprioception for In-Hand Dexterous Manipulation
Researchers have developed a transformer-based multimodal framework named NoContactNoWorries that allows robots to infer binary contact states from RGB-D vision and the robot's proprioception. This approach offers a…
OpenThoughts-Agent: Data Recipes for Agentic Models
The OpenThoughts-Agent project presents a fully open data curation pipeline for training broadly capable AI agents. The researchers conducted ablation experiments and used the pipeline to create a dataset of 100K…
OpenThoughts-Agent: Data Recipes for Agentic Models
The OpenThoughts-Agent project presents a fully open data curation pipeline for training agentic models, which are language models that expand AI applications. The project conducted controlled ablation experiments and…
Paying to Know: Micro-Transaction Markets for Verified Product Information in Agentic E-Commerce
The authors propose a new approach for e-commerce where buyers can pay small amounts of money to access detailed, verified product information through micro-transaction markets. This system aims to reward genuine…
Paying to Know: Micro-Transaction Markets for Verified Product Information in Agentic E-Commerce
The authors propose a new model for e-commerce where buyers pay small amounts of money to access detailed, verified information about products through the use of agent-native micro-payment rails. This model is designed…
Poster: Exploring the Limits of Audio-Based Detection of Turkish Phone Call Scams
Researchers introduce a new dataset of Turkish phone call conversations to test large language models' ability to detect scams. They find that using corrected or uncorrected transcripts improves results compared to…
SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation
Researchers have developed SAFARI, a framework that helps diagnose failures in autonomous agents by using a specialized toolbox for reading and searching trajectory segments and a persistent Short-Term Memory (STM) for…
Scaling Laws for Task-Specific LLM Distillation
This research provides empirical scaling laws for compressing Large Language Models (LLMs) specifically for certain domains, focusing on quantitative finance as the application domain. The study investigates the impact…
The Latent Bridge: A Continuous Slow-Fast Channel for Real-Time Game Agents
Researchers propose a new method called the Latent Bridge for real-time agents that can act quickly while still planning effectively, even under time constraints such as those found in video games. The method involves…
Visualizing 'We the People': Bridging the Perception Gap through Pluralistic Data Storytelling
The paper discusses a project that used AI to create interactive 'opinion landscapes' for an online deliberation called 'We the People'. The goal was to provide a more nuanced and collaborative way of visualizing…
World Models in Pieces: Structural Certification for General Agents
The paper proposes a method to evaluate the understanding of an AI agent in a complex world model, focusing on situations where the agent's abilities are specialized. The authors argue that standard guarantees fail to…
AI Exposure Scores: what they measure, what they miss, and what comes next
The article discusses the limitations of a set of AI Exposure Scores introduced by Eloundou et al., which measure the share of tasks that can be assisted by large language models. The authors highlight two main gaps…
AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction
The paper presents the Android Open Harness Project (AOHP), an operating system level agent harness built on the Android Open Source Project. AOHP aims to treat AI agents as first-class OS actors, enabling adaptive…
Detecting Malicious Agent Skills in the Wild using Attention
The research presents a two-stage detector called Locate-and-Judge for identifying malicious skills in Language Model (LLM) agents. These skills are third-party packages of instructions that can potentially exfiltrate…
GIF: Locally Sound Geometric Information Flow Control for LLMs
Researchers have proposed a new method called Geometric Information Flow (GIF) for controlling information flow in large language models used in agentic systems. GIF uses the model's Jacobian and local output geometry…
HOLMES: Evaluating Higher-Order Logical Reasoning in LLMs
The paper introduces HOLMES, a new benchmark for evaluating higher-order symbolic reasoning in Large Language Models (LLMs). This benchmark is designed to test the models' ability to reason over rules, predicates…
Kamera: Unified Position-Invariant Multimodal KV Cache for Training-Free Reuse
Researchers propose a new method called Kamera for efficient use of KV cache by multimodal agents during reasoning iterations. The method avoids the need to re-encode from scratch and reduces the need for multiple…
Learning Process Rewards via Success Visitation Matching for Efficient RL
Researchers propose a method to make reinforcement learning (RL) more effective when dealing with sparse rewards, which are typically given only when a task is completed. The method incentivizes the RL-learned policy…
Litmus: Zero-Label, Code-Driven Metric Specification for Evaluating AI Systems
The paper presents Litmus, a system that creates evaluation metrics for AI pipelines by understanding the purpose of the code and asking questions about it. It was tested on three real AI pipelines and outperformed…
POTracker: Optimizing Large Language Models for Standard-Compliant Power Outage Report Generation
Researchers have developed POTracker, an optimized large language model (LLM) for generating power outage reports that follow strict formatting and structural rules required by energy-sector regulatory bodies in the…
PsyBridge: A Hybrid Intelligent Framework for Multi-Dimensional Mental Health Assessment and Decision Support
The study introduces PsyBridge, a hybrid intelligent framework aimed at improving the multi-dimensional assessment of mental health by integrating clinically validated screening tools, cognitive evaluations, and…
RECALL: Recovery Experience Collection for Active Lifelong Learning in Vision-Language-Action Models
The paper proposes a new approach for training Vision-Language-Action (VLA) models through active, continual learning. This method aims to address the inefficiencies of traditional passive imitation learning by…
RS-Gen: A Multi-Stage Agentic Framework for Reasoning and Search-Augmented Image Generation
The paper introduces RS-Gen, a multi-stage image agent framework that aims to improve the performance of existing image generation and editing models by introducing a 'Questioning-and-Solving' closed-loop mechanism for…
Semantic Browsing: Controllable Diversity for Image Generation
Researchers have developed a new method for generating diverse images using text-to-image models. Instead of relying on stochastic variations within the model, they induce diversity directly at the text level. The…
Teaching LLMs String Matching, Backtracking, and Error Recovery to Deduce Bases and Truth Tables for the Combinatorially Exploding Bit Manipulation Puzzles
The paper presents a new approach to solve the NVIDIA Nemotron Model Reasoning Challenge's Bit Manipulation Puzzles, where the goal is to discover hidden rules for transforming binary strings. The approach avoids…
VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct
The paper presents VeriEvol, a framework for scaling reinforcement learning in visual mathematical reasoning. It addresses the challenge of maintaining reliable reward labels as data volume grows by treating scaling as…
VideoAgent: All-in-One Framework for Video Understanding and Editing
Researchers have developed a new framework called VideoAgent to automate video editing tasks. The system can handle diverse editing operations and understand long videos, addressing limitations in existing automated…
MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop
Researchers have developed a new benchmark for AI agents on macOS desktops called MacAgentBench. The benchmark includes 676 tasks across 25 applications and measures the performance of various models using a…
PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement
The paper introduces PAPERCLAW, a multi-agent system that automates the research process from start to finish. It uses a large language model as its core, which can drive the entire pipeline and also allows for human…
