arXiv||1 min read
Efficient SWE Agent Benchmarking via Trajectory-Aware Evaluation
Focuses on Efficient SWE Agent Benchmarking via Trajectory-Aware Evaluation.
Topic archive
Reasoning AI news
Evaluation, inference quality, and model cognition. This page collects the latest briefings that match the topic so readers can follow one area without scanning the full feed.
Indexed briefings
619
Latest source-linked updates, ordered newest first.
Latest
Recent briefings
arXiv||1 min read
TAG-Bench: Benchmarking Temporal Audio Grounding in Large Audio Language Models
Focuses on TAG-Bench: Benchmarking Temporal Audio Grounding in Large Audio Language Models.
arXiv||1 min read
EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation
Focuses on EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation.
arXiv||1 min read
Context-Aware Interleaved Batching for WhisperX
Focuses on Context-Aware Interleaved Batching for WhisperX.
arXiv||1 min read
Aspire: Can Models Self-Evolve from Vague Goals?
Focuses on Aspire: Can Models Self-Evolve from Vague Goals?.
arXiv||1 min read
Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents
Focuses on Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents.
arXiv||1 min read
S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?
Focuses on S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?.
arXiv||1 min read
Parsing the Stream: A Live Trace Model for Long-Horizon Agents and Their Observers
Focuses on Parsing the Stream: A Live Trace Model for Long-Horizon Agents and Their Observers.
arXiv||1 min read
Token-Efficient Data Reasoning Agents via Adaptive Structuring of Unstructured Data
Focuses on Token-Efficient Data Reasoning Agents via Adaptive Structuring of Unstructured Data.
arXiv||1 min read
Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence
Focuses on Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence.
arXiv||1 min read
EDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systems
Focuses on EDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systems.
arXiv||1 min read
LEAP: Likelihood Elicitation and Aggregation for LLM-based Probabilistic Forecasting
Focuses on LEAP: Likelihood Elicitation and Aggregation for LLM-based Probabilistic Forecasting.
arXiv||1 min read
LLM-Based Agents for Software and Systems Security: Approaches, Applications, and Assessment
Focuses on LLM-Based Agents for Software and Systems Security: Approaches, Applications, and Assessment.
arXiv||1 min read
Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols
Focuses on Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols.
arXiv||1 min read
COVER: Identifiable Evaluation of Coalition Routing
Focuses on COVER: Identifiable Evaluation of Coalition Routing.
arXiv||1 min read
Explore Before Committing: Hypothesis-Guided Search for Deep Research Agents
Focuses on Explore Before Committing: Hypothesis-Guided Search for Deep Research Agents.
arXiv||1 min read
A Human-in-the-Loop Autonomous Agent for Industry Time Series Forecasting
Focuses on A Human-in-the-Loop Autonomous Agent for Industry Time Series Forecasting.
arXiv||1 min read
Learning to Use Tools: Reinforcement Learning for Tool-Integrated Mathematical Reasoning
Focuses on Learning to Use Tools: Reinforcement Learning for Tool-Integrated Mathematical Reasoning.
arXiv||1 min read
When Verified Source Becomes Attack Input: Defending Smart Contracts Against LLM-Based Vulnerability Scanning
Focuses on When Verified Source Becomes Attack Input: Defending Smart Contracts Against LLM-Based Vulnerability Scanning.
arXiv||1 min read
RetailAgent: Structured Adverse Timing in Self-Conditioned Multimodal LLM Trading Agents
Focuses on RetailAgent: Structured Adverse Timing in Self-Conditioned Multimodal LLM Trading Agents.
arXiv||1 min read
PersonaForge: Realistic Multi-Turn User Simulation for Agentic Systems
Focuses on PersonaForge: Realistic Multi-Turn User Simulation for Agentic Systems.
arXiv||1 min read
TRIPPULSE: Multi-Agent Travel Planning with Review-Grounded Reasoning
Focuses on TRIPPULSE: Multi-Agent Travel Planning with Review-Grounded Reasoning.
arXiv||1 min read
Continuous Autonomous Refactoring: A Research Roadmap for AI-Driven Code Quality Maintenance
Focuses on Continuous Autonomous Refactoring: A Research Roadmap for AI-Driven Code Quality Maintenance.
arXiv||1 min read
EvoUndo: Recoverability-Constrained Self-Evolution for LLM Agent Harnesses
Focuses on EvoUndo: Recoverability-Constrained Self-Evolution for LLM Agent Harnesses.
arXiv||1 min read
AGENT-O: A Semantic Agent Card Framework for Interoperable and Governed Healthcare AI Agents
Focuses on AGENT-O: A Semantic Agent Card Framework for Interoperable and Governed Healthcare AI Agents.
arXiv||1 min read
LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering
Focuses on LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering.
arXiv||1 min read
STEGNav: Spatio-Temporal Event Graph Reasoning for Multimodal Lifelong Object Navigation
Focuses on STEGNav: Spatio-Temporal Event Graph Reasoning for Multimodal Lifelong Object Navigation.
arXiv||1 min read
Parser States Already Know: Structure-Conditioned KV Persistence for Structured Generation
Focuses on Parser States Already Know: Structure-Conditioned KV Persistence for Structured Generation.
arXiv||1 min read
CoCoBench: A Cooperative Coordination Benchmark for Embodied Multi-Agent Task Planning
Focuses on CoCoBench: A Cooperative Coordination Benchmark for Embodied Multi-Agent Task Planning.
arXiv||1 min read
Finding Where the Buck Stops: An Automated Failure Attribution-Based Reflection Framework for Multi-Agent Collaboration
Focuses on Finding Where the Buck Stops: An Automated Failure Attribution-Based Reflection Framework for Multi-Agent Collaboration.
arXiv||1 min read
Beyond Task-Only Matching: Personalized Skill Routing with Counterfactual Evaluation
Focuses on Beyond Task-Only Matching: Personalized Skill Routing with Counterfactual Evaluation.
arXiv||1 min read
Schwarz: Solver-Aware Agentic Program Verification
Focuses on Schwarz: Solver-Aware Agentic Program Verification.
arXiv||1 min read
Adaptive Strategy Generation for Boundary Value Exploration Beyond Numeric Inputs
Focuses on Adaptive Strategy Generation for Boundary Value Exploration Beyond Numeric Inputs.
arXiv||1 min read
MuSP-Bench: Advanced Multimodal Benchmarking of Music Understanding across Score and Performance
Focuses on MuSP-Bench: Advanced Multimodal Benchmarking of Music Understanding across Score and Performance.
arXiv||1 min read
Benchmarking large language model agent societies against human behavioural distributions
Focuses on Benchmarking large language model agent societies against human behavioural distributions.
arXiv||1 min read
RedEvoAgent: Automatic Red-Teaming Agent with Experience-Driven Skill Evolution
Focuses on RedEvoAgent: Automatic Red-Teaming Agent with Experience-Driven Skill Evolution.