arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2502.03438cs.AI
收藏

BFS-Prover:基于LLM的自动定理证明的可扩展最佳优先树搜索

BFS-Prover: Scalable Best-First Tree Search for LLM-based Automatic Theorem Proving

ByteDance Seed(字节跳动种子);ByteDance Applied Machine Learning(字节跳动应用机器学习);Stanford University(斯坦福大学)

Ran Xin, Chenguang Xi, Jie Yang, Feng Chen, Hang Wu, Xia Xiao, Yifan Sun, Shen Zheng, Kai Shen

AI总结:

本文提出BFS-Prover,一种可扩展的最佳优先树搜索框架,通过数据过滤、DPO优化和长度归一化,在MiniF2F上达到72.95%的最先进性能,证明简单方法可媲美复杂搜索。

摘要:

近年来,大型语言模型(LLM)的进步激发了使用Lean4进行自动定理证明的日益增长的兴趣,其中有效的树搜索方法对于在庞大的证明搜索空间中导航至关重要。虽然现有方法主要依赖于价值函数和/或蒙特卡洛树搜索(MCTS),但像最佳优先树搜索(BFS)这样更简单的方法的潜力仍未得到充分探索。在本文中,我们研究了BFS是否能在大规模定理证明任务中实现具有竞争力的性能。我们提出了BFS-Prover,一个可扩展的专家迭代框架,具有三个关键创新。首先,我们在每轮专家迭代中实施策略性数据过滤,排除可通过束搜索节点扩展解决的问题,以专注于更难的案例。其次,我们通过直接偏好优化(DPO)应用于自动注释了编译器错误反馈的状态-策略对,来提高BFS的样本效率,优化LLM的策略以优先进行富有成效的扩展。第三,我们在BFS中采用长度归一化,以鼓励探索更深的证明路径。BFS-Prover在MiniF2F测试集上达到了最先进的72.95%的分数,因此挑战了复杂树搜索方法的必要性,证明了BFS在适当扩展时可以实现具有竞争力的性能。为了促进该领域的进一步研究和发展,我们已在https://huggingface.co/ByteDance-Seed/BFS-Prover-V1-7B开源了我们的模型。

英文摘要:

Recent advancements in large language models (LLMs) have spurred growing interest in automatic theorem proving using Lean4, where effective tree search methods are crucial for navigating the underlying large proof search spaces. While the existing approaches primarily rely on value functions and/or Monte Carlo Tree Search (MCTS), the potential of simpler methods like Best-First Tree Search (BFS) remains underexplored. In this paper, we investigate whether BFS can achieve competitive performance in large-scale theorem proving tasks. We present BFS-Prover, a scalable expert iteration framework, featuring three key innovations. First, we implement strategic data filtering at each expert iteration round, excluding problems solvable via beam search node expansion to focus on harder cases. Second, we improve the sample efficiency of BFS through Direct Preference Optimization (DPO) applied to state-tactic pairs automatically annotated with compiler error feedback, refining the LLM's policy to prioritize productive expansions. Third, we employ length normalization in BFS to encourage exploration of deeper proof paths. BFS-Prover achieves a state-of-the-art score of $72.95\%$ on the MiniF2F test set and therefore challenges the perceived necessity of complex tree search methods, demonstrating that BFS can achieve competitive performance when properly scaled. To facilitate further research and development in this area, we have open-sourced our model at https://huggingface.co/ByteDance-Seed/BFS-Prover-V1-7B.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑