01. Who Holds the Pen? Let Specifications, Not Agents, Sign Off

85% worth reading · novelty 2.0/3 · new method · cs.MA

02. ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds

85% worth reading · novelty 2.0/3 · benchmark/eval · cs.AI

03. Coding Agents for Generalized Task and Motion Planning Problems

84% worth reading · novelty 1.9/3 · benchmark/eval · cs.AI

04. JevOut: Natural Context Can Flip Decision Models

87% worth reading · novelty 1.6/3 · benchmark/eval · cs.CL

05. LLM Agents Can Easily Tamper With Their Own Traces

84% worth reading · novelty 1.8/3 · benchmark/eval · cs.AI

06. Instrumental Monitor Evasion Emerges Under Ordinary Task Pressure

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

07. Agentic Detection of Online Conspiracies

83% worth reading · novelty 1.8/3 · new method · cs.CL

08. RAPID: Robot Agentic Programming from Demonstrations

81% worth reading · novelty 1.8/3 · new method · cs.AI

09. Screen Before You Serve: Simulation for Production Customer Experience AI Agents at 140M Scale

86% worth reading · novelty 1.5/3 · applied system · cs.CL

10. Codetta: High-Capacity, Keyless, and Undetectable Multi-Agent Collusion

78% worth reading · novelty 2.0/3 · new method · cs.MA

11. Multimodal Thinking with Renderable Programs

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

12. Progressive Skill Discovery as Access Control for Tool-Using LLM Agents: Structural Governance through Role-Scoped Capability Delivery

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

13. Augur: A Synthetic Decision Lab for Rehearsing Reactions to Product and Policy Changes

78% worth reading · novelty 1.9/3 · benchmark/eval · cs.MA

14. What, When, and How: Audio Description as Constrained Global Optimization

76% worth reading · novelty 1.9/3 · new method · cs.CL

15. AlphaDiverse: Post-Training Local Quantitative Research Agents for Diverse Exploration in Alpha Factor Mining

78% worth reading · novelty 1.6/3 · new method · cs.MA

16. GRASP: Generating, Revising, and Assessing for Strategic Planning with Agentic AI

78% worth reading · novelty 1.5/3 · new method · cs.CL

17. ARGUS: Role-Aware Event Knowledge Graphs for U.S. Employment-Discrimination Complaints

79% worth reading · novelty 1.3/3 · applied system · cs.CL

18. REAT: A Reflective Experience-Augmented Tutoring Framework for Multi-turn Mathematical Instruction

75% worth reading · novelty 1.6/3 · new method · cs.MA

19. SAGE: Mitigating Long-Horizon Reasoning Biases via Topological Guidance

67% worth reading · novelty 2.0/3 · new method · cs.AI

20. Compliant with Local Controls, Collectively Discriminatory. A Governance Architecture for Multi-Agent AI in Regulated Finance

60% worth reading · novelty 2.0/3 · new method · cs.MA