Fetching the paper…

The Surprising Effectiveness of Negative Reinforcement in LLM Reasoning · Around