Unleashing the Reasoning Potential of Pre-trained LLMs by Critique Fine-Tuning on One Problem

Open in new window