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 39 of 59
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…
When Synthetic Speech Is All You Have: Better Call GRPO
The paper presents a new approach for adapting Large Language Model (LLM) based Automatic Speech Recognition (ASR) systems to regulated domains like banking using synthetic text-to-speech instead of real speech due to…
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…
Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows
The paper proposes a conceptual model for large language model (LLM) applications that use explicit workflows. The model is designed to represent various elements of the workflow such as definitions, instances, and…
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 a new method called Agon for reinforcement learning that allows two competing models to grade each other's reasoning during training. This approach aims to improve the quality of solutions by…
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…
Breaking Database Lock-in: Agentic Regeneration of High Performance Storage Readers for Database Bypass
The researchers present an approach called Jailbreak that allows analytical workloads to bypass the traditional database driver and read storage files directly, using Large Language Models (LLMs) to regenerate table…
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…
Co-LMLM: Continuous-Query Limited Memory Language Models
Researchers have developed a new type of language model called Co-LMLM (Continuous-Query Limited Memory Language Model). This model stores factual knowledge in a knowledge base instead of memorizing it in its weights…
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…
DiaLLM: An Investigation into the Robustness-Generation Gap in English Dialect Adaptation
Researchers introduce DiaLLM, a system that aims to address the issue of large language models only producing standard English while understanding dialectal English. The study compares three open-weight language model…
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…
From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
Researchers propose a framework called EvoSOP to help large language model (LLM) agents optimize their toolset by synthesizing atomic actions into reusable Standard Operating Procedures (SOPs). This approach aims to…
Future Confidence Distillation in Large Language Models
In this research paper from 2026, Sahil Kale explores the importance of accurate confidence estimation in large language models (LLMs) for various applications like retrieval and tool use. The study focuses on how…
Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety
The paper introduces institutional red-teaming, a method for evaluating deployment rules in multi-agent AI. The study uses IABench-CA, a consequence-allocation benchmark, to test five canonical rules across seven model…
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…
ORCAID: Oblique Rule-Based Continuous-Action Interpretation for Deep RL Policies
Researchers introduce ORCAID, a method for extracting interpretable rule-based policies from reinforcement learning (RL) agents in complex environments with continuous action spaces. The method uses an efficient…
Physics-Audited Agentic Discovery in Scientific Machine Learning
Researchers have developed Physics-Audited Agentic Scientific Machine Learning (PA-SciML), a new workflow for machine learning in science that ensures discovered models follow certain physics rules such as boundary…
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…
QCNN with Rough Path Signature Kernels
The research paper proposes a hybrid quantum-classical architecture to address time series classification challenges, which are common in various scientific and engineering domains. The architecture integrates Quantum…
Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
Researchers found that LLM agents (Language Model-based tools) can violate policies without being detected because the resulting state changes are not reported by the tools or the agent. The researchers studied this…
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…
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.
