Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
Yue et al., 2025, Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model? (arXiv preprint arXiv:2504.13837)
Subscriber deck — free during early access
1 cards in “4.5. Effects of Entropy” · whole deck
4.5. Effects of Entropy
- Is the narrowing just lower output entropy after RL?Only partly. Raising the RL model’s sampling temperature until its entropy matched the base model’s improved its large-\(k\) pass@k somewhat, but it still fell below the base model.
- Yue et al., 2025, Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model? (arXiv preprint arXiv:2504.13837), 4.5. Effects of Entropy
Source
- Yang Yue, Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang, Yang Yue, Shiji Song and Gao Huang. 2025. Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?. In arXiv preprint arXiv:2504.13837. https://arxiv.org/abs/2504.13837