LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget Paper • 2607.14952 • Published 3 days ago • 148
SWE-Review: Closing the Loop on Issue Resolution with Agentic Code Review Paper • 2607.06065 • Published 12 days ago • 8
The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning Paper • 2606.29526 • Published 21 days ago • 167
Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios Paper • 2605.28618 • Published May 27 • 32
Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism Paper • 2606.00408 • Published May 29 • 65
Representation over Routing: Diagnosing Temporal Routing Pathologies in Multi-Timescale PPO Paper • 2604.13517 • Published May 30 • 5
IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools Paper • 2605.20682 • Published May 20 • 85