01.Free to Look, Free to Reason: Simple LLM Harnesses Outdiagnose Hand-Crafted Root Cause Analysis Agents

85% worth reading · novelty 1.7/3 ·benchmark/eval ·cs.AI

02.I-HITL: Inalienable Human-in-the-Loop

80% worth reading · novelty 2.0/3 ·new method ·cs.AI

03.Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models

79% worth reading · novelty 1.9/3 ·new method ·cs.AI

04.Quantifying Claim Smoothing by LLM Research Agents

81% worth reading · novelty 1.8/3 ·benchmark/eval ·cs.CL

05.ClinVerse: A Realistic Multi-Agent Diagnostic Environment for Training and Evaluating Medical Agents

83% worth reading · novelty 1.6/3 ·benchmark/eval ·cs.MA

06.Pretraining Knows What, Simulation Teaches How: Compositional Generalization in a Sim-Trained Robot Agent

82% worth reading · novelty 1.7/3 ·new method ·cs.AI

07.Knowing When to Escalate: Disagreement Patterns for Uncertainty-Aware Content Moderation

79% worth reading · novelty 1.9/3 ·new method ·cs.MA

08.Work Agents: Systems, Capabilities, and Optimization

84% worth reading · novelty 1.5/3 ·survey/position ·cs.LG

09.Collective Emergent Alignment: Measuring and Amplifying Interaction-Induced Reasoning Alignment in Multi-Agent LLMs

80% worth reading · novelty 1.8/3 ·new method ·cs.MA

10.EvoRouter: Cost-Efficient Multi-Agent Routing via Insight-Guided On-Policy Distillation

80% worth reading · novelty 1.7/3 ·new method ·cs.MA

11.SlowLab: Benchmarking Autonomous AI Agents on a Year of Experiments They Cannot Undo

75% worth reading · novelty 1.9/3 ·benchmark/eval ·cs.MA

12.FOCUS: Active Critic Learning for Long-Horizon Language Model Agents

80% worth reading · novelty 1.6/3 ·new method ·cs.LG

13.MiRL: Error-Targeted Reinforcement Learning for Correcting Terminal Agent Misbehaviors

81% worth reading · novelty 1.4/3 ·new method ·cs.CL

14.AgentProofBench: Benchmarking Reusable Verifiers for Web Agents

75% worth reading · novelty 1.7/3 ·benchmark/eval ·cs.MA

15.Agentic AI Orchestration Framework for Spatial Intelligence: Supply Chain Deployment

76% worth reading · novelty 1.5/3 ·applied system ·cs.MA

16.Retriever-Free Retrieval: Navigating Documents with Decision Models

68% worth reading · novelty 2.0/3 ·new method ·cs.AI

17.Auto-Research Arena: Benchmarking and Advancing Auto-Research in Language Modeling

70% worth reading · novelty 1.9/3 ·benchmark/eval ·cs.LG

18.A Unified Arena for LLM Agents across Diverse Eurogame

73% worth reading · novelty 1.6/3 ·benchmark/eval ·cs.AI

19.Profiling a Frontier Agent with EBench: From Direct Control to Harness with Policy

80% worth reading · novelty 1.0/3 ·benchmark/eval ·cs.AI

20.GEAR: Feature-Based Configuration and Code Generation for Agentic Systems

71% worth reading · novelty 1.5/3 ·applied system ·cs.MA