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 35 of 59
From Critic to Confidence: PPO for Language-Based Quantitative Prediction with Confidence Estimation
Researchers introduce a new framework called CARE-PPO to improve language-based quantitative prediction by both accurately estimating numerical values and providing reliable confidence signals. The method uses…
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…
Knowledge- and Gradient-Guided Reinforcement Learning for Parametrized Action Markov Decision Processes
In this research paper, the authors present a novel reinforcement learning algorithm called Knowledge- and Gradient-Guided Reinforcement Learning (KGRL). This algorithm is designed to work in Parametrized Action Markov…
Less Experts, Faster Decoding: Cost-Aware Speculative Decoding for Mixture-of-Experts
Researchers propose a new method called EcoSpec for optimizing the performance of Large Language Models (LLMs) using Sparse Mixture-of-Experts (MoE). This method aims to improve decoding speed by considering the cost…
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.
LLMs Can See the Smoke but not the Fire: Evaluating Abductive Reasoning with Elenchos
Researchers introduce Elenchos, a framework to evaluate abductive reasoning (inferring latent hypotheses) of Large Language Models (LLMs). The study finds that LLMs can often detect changes in formal systems but…
MemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversations
The paper introduces MemOps, a new benchmark for evaluating the memory capabilities of large language models in extended, multi-session interactions. Unlike existing benchmarks that focus on downstream question…
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…
PalmClaw: A Native On-Device Agent Framework for Mobile Phones
The paper introduces PalmClaw, a native agent framework designed to run on mobile phones. It aims to overcome the limitations of existing mobile agents by directly accessing device capabilities, making each action…
Practical Judgment, Virtue, and Intuition in the Use of Opaque AI-Enabled Systems
The article discusses the challenges of using AI-enabled systems with opaque functions in various domains, such as military applications. It suggests that leveraging human abilities like practical judgment, virtue, and…
Reproducible Reservoir Computing with Thermally Driven Superparamagnets: Controlling Temperature Sensitivity
The authors propose a new type of computing system using superparamagnetic nanodots, which are sensitive to temperature fluctuations due to their thermal activation dynamics. In this paper, they simulate how…
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…
TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale
TerraZero is a procedural driving simulator and training stack for reinforcement learning at scale, designed to train autonomous driving agents without human demonstrations or fallback planners. The simulator provides…
Tracing Agentic Failure from the Flow of Success
The text discusses a new model called OAT for failure attribution in LLM-based agentic systems. The model aims to identify the steps causing failures in these systems, which is crucial for debugging and improvement…
Unveiling Complex Collective Behaviors from Simple Rewards
The paper proposes a method called the Agent Response Map (ARM) to analyze Multi-agent Reinforcement Learning (MARL) and understand the mechanisms behind complex behaviors that emerge from simple rewards. The ARM…
UR-VC: Unsupervised Robotic Value Correction for Time-Derived Progress Proxies
Researchers introduce an offline method called Unsupervised Robotic Value Correction (UR-VC) to correct time-derived progress labels used by modern robot learning systems. The method is based on a regularity in…
Vertical Standardisation for High-Risk AI Systems under the EU AI Act: A Domain-Specific Framework for Algorithmic Hiring
The authors propose a domain-specific framework for standardizing high-risk AI systems, particularly in recruitment, in compliance with the EU AI Act. The framework focuses on areas like risk management, data quality…
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 Glimpse into Long-term Physical Coexistence with Intelligent Robots
The paper introduces PHILIA, a multi-robot agent designed to facilitate long-term physical coexistence with intelligent robots. The agent aims to support diverse user interfaces, maintain long-horizon memory of users…
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.
