TongSIM: A General Platform for Simulating Intelligent Machines Paper • 2512.20206 • Published 4 days ago • 26
Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning Paper • 2512.07461 • Published 19 days ago • 74
Understanding and Leveraging the Expert Specialization of Context Faithfulness in Mixture-of-Experts LLMs Paper • 2508.19594 • Published Aug 27 • 2