Field notes
2,343 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.
262 of 2,343 notes match.
- Question answered
- Source note
- Topic
- Shared notes
Notes on artificial intelligence
262 notes, page 1 of 11
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…
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…
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…
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…
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…
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…
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…
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…
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…
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…
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…
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…
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…
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…
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
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
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…
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.
