learn.ifs.name

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Shao et al., 2024, DeepSeekMath (arXiv preprint arXiv:2402.03300)

Subscriber deck — free during early access

1 cards in “5.2.2 Why RL Works?” · whole deck

5.2.2 Why RL Works?

  1. Why does RL improve DeepSeekMath, according to its Pass@K and Maj@K analysis?
    RL raised Maj@K (majority-vote accuracy) but not Pass@K (whether any of K samples is right): it made the output distribution more robust by boosting correct answers already among the top samples, rather than adding fundamentally new capabilities.
    • Shao et al., 2024, DeepSeekMath (arXiv preprint arXiv:2402.03300), §5.2.2

Source

  • Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y. K. Li, Y. Wu and Daya Guo. 2024. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models. In arXiv preprint arXiv:2402.03300. https://arxiv.org/abs/2402.03300