More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning Paper • 2610.19789 • Published 8 days ago • 93
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published 22 days ago • 644
FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis Paper • 2608.18580 • Published Aug 19 • 75