TongSIM: A General Platform for Simulating Intelligent Machines Paper • 2512.20206 • Published 11 days ago • 27
Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning Paper • 2512.07461 • Published 26 days ago • 74
Reverse-Engineered Reasoning for Open-Ended Generation Paper • 2509.06160 • Published Sep 7, 2025 • 150
Understanding and Leveraging the Expert Specialization of Context Faithfulness in Mixture-of-Experts LLMs Paper • 2508.19594 • Published Aug 27, 2025 • 2