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 2602.10538stat.MLcs.LG
收藏

为什么智能体定理证明器有效:数学推理模型的统计可证明性理论

Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models

CyberAgent, Inc.(CyberAgent公司);National Institute of Informatics, Japan(日本信息机构)

Sho Sonoda, Shunta Akiyama, Yuya Uezato

AI总结:

本文通过统计可证明性理论,将形式化证明搜索建模为有限视界可达性MDP,分析了智能体定理证明器中各组件对有限预算下证明成功率的影响,并给出了成功率差距的误差界。

摘要:

智能体定理证明器结合了推理模型、检索、搜索和证明助手验证器,但目前尚不清楚哪些组件实际上提高了有限预算下的证明成功率,以及它们为何在真实数学工作负载上有效。我们通过统计可证明性来研究这个问题:在指定定理实例流上,在预算内达到已验证证明的概率。我们将形式化证明搜索建模为具有确定性验证器动态的有限视界可达性MDP,并表明在忠实状态抽象下,最优成功概率与普通句法可证明性一致。然后我们分析了一个简单但实际重要的流程:深度上的离线动作值回归,随后是贪婪的测试时证明。我们的主要定理通过一个占用加权和的一致动作值误差来界定学习证明器与最优证明器之间的可证明性差距;在常见的均匀误差解读中,主要复杂度乘子是学习证明器的平均截断证明长度。误差分解为逼近误差、训练分布的几何覆盖和蒙特卡洛标签噪声,并在动作间隔边界条件下以快速速率改进。该结果给出了一个组件敏感的解释,说明为什么验证器反馈、检索、表示几何和证明缩短机制在偏置定理工作负载上有帮助,而不与经典的最坏情况困难性相矛盾。

英文摘要:

Agentic theorem provers combine a reasoning model, retrieval, search, and a proof assistant verifier, yet it remains unclear which components actually improve finite-budget proof success and why they help on real mathematical workloads. We study this question through statistical provability: the probability of reaching a verified proof within a budget on a specified stream of theorem instances. We model formal proof search as a finite-horizon reachability MDP with deterministic verifier dynamics, and show that under a faithful state abstraction the optimal success probability coincides with ordinary syntactic provability. We then analyze a simple but practically important pipeline: depth-wise offline action-value regression followed by greedy test-time proving. Our main theorem bounds the provability gap between the learned prover and the optimal prover by an occupancy-weighted sum of uniform action-value errors; in the common uniform-error reading, the leading complexity multiplier is the learned prover's average truncated proof length. The error decomposes into approximation error, geometric coverage of the training distribution, and Monte Carlo label noise, and improves to a fast rate under an action-gap margin condition. The result gives a component-sensitive account of why verifier feedback, retrieval, representation geometry, and proof-shortening mechanisms help on biased theorem workloads, without contradicting classical worst-case hardness.

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。

↑