PapersMulti-source

Turing Agent:用智能体工作流形式化验证研究级数学证明

Turing Agent: Agentic Formal Verification for Mathematical Research

Why we picked it

arXiv 新论文,一个专门做数学证明形式化验证的智能体框架,底层用 GLM-5.3 和 DeepSeek-V4-Pro,跑赢了 Codex CLI 和 Claude Code,还解决了 OpenAI 问题集里三道没被啃下来的理论计算机难题。做形式化验证或 Lean 相关工作的开发者值得看看它的双层架构怎么拆任务。

论文提出 Turing Agent,一个面向研究级数学端到端形式化验证的智能体框架,采用定理形式化与证明形式化双层架构,结合结构化证明分解、并行验证、证明经验积累和选择性语义审查。在理论计算机科学(TCS)方向的评估中,基于 GLM-5.3 和 DeepSeek-V4-Pro,Turing Agent 在 Formal TCS Bench 上达到约 57-62% 的忠实形式化成功率,在 RLMEval 的 TCS 子集上取得 83% 的整体 pass@1。对比对象包括直接生成以及 Codex CLI 和 Claude Code 这类通用编程智能体框架。更关键的成果是:基于 DeepSeek 构建的 Turing Agent 完成了 OpenAI 2026 年 10 月问题集中三个此前未被解决的 TCS 任务的形式化验证,其中一道难题需要约一百条经过形式化验证的辅助命题。

Body

技术机制上,该框架把证明验证拆解为可并行的小任务:先由定理形式化智能体把数学陈述转成 Lean 等形式语言中的精确定义,再由证明形式化智能体逐条验证分解后的子命题,并积累经验供后续复用。评估刻意选择 TCS 作为测试领域,因为该领域涉及复杂的量词嵌套、依赖关系和严格的语义要求,是形式化验证中最困难的一类问题。

对研究者和工程团队的意义在于:随着 LLM 生成证明数量激增,机器可验证的正确性检查正成为瓶颈,Turing Agent 展示了智能体编排可以把验证成功率显著推高,同时能在验证过程中发现更简化的替代证明、暴露原证明中不必要的依赖,产出可复用的数学知识。

Original · arXiv cs.AI

Turing Agent: Agentic Formal Verification for Mathematical Research

Large language models (LLMs) are rapidly advancing mathematical discovery, moving beyond benchmark-level problem solving to tackle frontier research problems. This progress is ushering mathematics into an era of proof abundance, raising a fundamental challenge: how can we ensure the correctness of the ever-growing volume of AI-generated proofs and transform them into trustworthy, machine-verified mathematical knowledge? We introduce Turing Agent, an agent harness designed to orchestrate end-to-end formal verification of research-level mathematics. Its wo-scale architecture integrates theorem formalization and proof formalization agents, empowered by structured proof decomposition, parallel verification, accumulated proof experience, and selective semantic review. Leveraging our collaborations with theoretical computer science (TCS) researchers, we focus our evaluation on TCS, a particularly demanding domain characterized by intricate quantification, complex dependencies, and stringent semantic requirements. Against direct generation and general-purpose coding harnesses, including Codex CLI and Claude Code, Turing Agent achieves approximately 57-62% faithful formalization success on Formal TCS Bench and 83% overall pass@1 on a TCS-focused subset of RLMEval, using GLM-5.3 and DeepSeek-V4-Pro. Notably, Turing Agent, built on DeepSeek, successfully completes the formal verification of three previously unresolved TCS tasks from OpenAI's October 2026 collection. One particularly challenging TCS problem required about a hundred formally verified auxiliary claims. Moreover, by actively exploring and refining the formalization process, Turing Agent can uncover simpler alternative proofs, expose unnecessary dependencies in original proofs, and generate reusable mathematical knowledge as a byproduct of verification.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Turing Agent, DeepSeek-V4-Pro, GLM-5.3, formal verification.

Source attribution: curated from arXiv cs.AI (@Yang Chen, Qianyi Wang, Ronghui Yang, Zhanhao Li, Yu Cong, Chen Zhang, Hangfan Zhang, Shao Zhang, Shuai Li, Toru Takisaka, Chao Xu, Yi Zhou, Zhihao Tang, Xiao Wang, Bo Zhang, Shuyue Hu). Cross-source corroboration: 11 independent sources have covered this development (including 掘金本周最热, AIGCLINK, lmarena.ai), confirming industry-wide signal strength.