Field notes
1,406 notes from my AI research vault, mapped by topic. Pick a topic to see its notes, or search.
Click the web, then scroll to zoom
Every note is in the list below. The web of topics draws here once the page has loaded.
1,406 notes and 240 topics.
- Question answered
- Source note
- Topic
- Shared notes
All notes, newest first
1,406 notes, page 43 of 59
Relational Multi-Agent Reinforcement Learning for Dynamic Pricing in High-Speed Railway Markets
Researchers propose an entity graph modeling approach for dynamic pricing in liberalized railway systems, using a modified multi-agent twin delayed deep deterministic policy gradient algorithm with graph-based…
Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation
The paper discusses a research project aimed at improving the capabilities of visual generators by addressing their limitation in fabricating what they don't know about the ever-evolving and vast visual world. The…
Selective Disclosure Watermarking for Large Language Models
Researchers have proposed a new watermarking method for large language models (LLMs) that allows selective disclosure of embedded metadata. This means that the method can control which parts of the embedded message are…
SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints
The text discusses a new benchmark called SovereignPA-Bench for evaluating the performance of user-owned personal agents. These agents are designed to remember preferences, filter information, use tools, and negotiate…
SPEARBench: A Benchmark for Naturalness Evaluation in Streaming Speech-to-Speech Language Models
A new benchmark called SPEARBench is introduced to evaluate the naturalness of speech-to-speech language models during question-answer interactions. The benchmark covers aspects like response latency, interruptions…
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…
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…
Unified Audio Intelligence Without Regressing on Text Intelligence
The researchers introduce a unified audio-text language model called Audex, which is built on a strong text-only model. Audex allows for strong fusion of audio and text, seamless multimodal generation, and…
Weak-to-Strong Generalization via Direct On-Policy Distillation
Researchers propose a method called Direct On-Policy Distillation (Direct-OPD) for improving stronger language models by leveraging the knowledge gained from reinforcement learning on smaller, weaker models. This…
When Claws Remember but Do Not Tell: Stealthy Memory Injection in Persistent Personal Agents
Researchers have developed a new type of attack called 'stealth memory injection' that targets persistent personal agents, which combine long-term memory and access to users' external environments. The attack involves…
Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses
Researchers introduce a new type of attack, called Forged Amplifying Rationale Memory Attack (FARMA), that targets the reasoning history of large language model (LLM) agents. FARMA inserts forged reasoning traces using…
A$^{2}$utoLPBench: An Auto-Generated, Agent-Friendly LP Benchmark via Inverse-KKT Construction
The paper introduces A$^{2}$utoLPBench, a new linear programming (LP) benchmark for testing LLM-driven agents. The benchmark generates LP problems in plain text, and uses a known feasible point and dual to write down a…
AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents
The paper introduces a new contract for long-horizon LLM agents that restricts the amount of past information available during decision-making in the game Slay the Spire 2. The new contract allows for easier isolation…
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…
Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials
The researchers compare the performance of various matrix-structured optimizers (including Muon, SOAP, and SOAP-Muon) to Adam in training machine learning interatomic potentials (MLIPs). They found that these…
Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support
The paper introduces Copewell, a novel multi-agent swarm system designed to improve mental health support, particularly for people in low- and middle-income countries where access is limited due to various factors. The…
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…
DemoPSD: Disagreement-Modulated Policy Self-Distillation
The paper introduces DemoPSD, a novel framework for training large language models (LLMs) that addresses the issues of overfitting, suppression of exploration, cross-domain generalization problems, and privileged…
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…
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…
An AI research agent drafts these notes from public papers and posts, and each one keeps its source link. Only the folders I choose to publish leave my Obsidian vault.
