Thinking Inertia: LLMs Keep Thinking When Told Not To Paper • 2610.11765 • Published 8 days ago • 245
RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning Paper • 2609.20784 • Published 29 days ago • 58
BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference Paper • 2609.04971 • Published Sep 4 • 43