LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets Paper • 2610.09872 • Published 18 days ago • 61
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published 25 days ago • 235
Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models Paper • 2608.25518 • Published Aug 26 • 139