False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published Sep 30 • 130
CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers? Paper • 2610.07557 • Published Oct 6 • 11
TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models Paper • 2610.07767 • Published Oct 6 • 43