Unleashing the Reasoning Potential of Pre-trained LLMs by Critique Fine-Tuning on One Problem Paper • 2506.03295 • Published Jun 3, 2025 • 17 • 2