Fetching the paper…

Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs · Around