PulseAugur
实时 12:31:54
English(EN) On Compositional Learning Behaviours in Formal Mathematics

AI代理形式化26本数学教科书,创建已验证的库

研究人员开发了AutoformBot,这是一个使用LLM和形式化验证工具将非正式数学散文翻译成机器可检查代码的多代理系统。该系统已应用于26本数学教科书,在Lean 4中创建了一个包含超过45,000个声明和500,000行代码的已验证库。另外,一项关于形式数学中组合学习行为的研究发现,虽然搜索密集型模型可以处理基本任务,但要达到奥林匹克竞赛水平的性能需要强大的组合学习能力,而这种能力对于复杂的数学验证来说是必要但不充分的。 AI

影响 AI形式化复杂数学推理能力方面的进步可能会加速AI驱动的科学发现和定理证明。

排序理由 两篇arXiv论文详细介绍了形式数学和AI方面的新方法和发现。

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

报道来源 [3]

  1. arXiv cs.AI TIER_1 English(EN) · Ahmad Rammal, Niket Patel, Fabian Gloeckle, Amaury Hayat, Julia Kempe, Remi Munos, Charles Arnal, Vivien Cabannes ·

    Formalizing Mathematics at Scale

    arXiv:2605.29955v1 Announce Type: new Abstract: We present AutoformBot, a multi-agent system for building an Autoformalized Textbook Library At Scale (Atlas) in Lean 4. AutoformBot orchestrates thousands of LLM agents, equipped with formal verification tools, dependency-aware tas…

  2. arXiv cs.CL TIER_1 English(EN) · Kevin Yandoka Denamgana\"i ·

    On Compositional Learning Behaviours in Formal Mathematics

    arXiv:2605.28512v1 Announce Type: new Abstract: Self-evolving scientific agents capable of conquering the hard tail of formal mathematics require Compositional Learning Behaviours (CLBs) -- the capacity to ground and recombine novel symbolic structures in context, beyond mere rec…

  3. arXiv cs.CL TIER_1 English(EN) · Kevin Yandoka Denamganaï ·

    On Compositional Learning Behaviours in Formal Mathematics

    Self-evolving scientific agents capable of conquering the hard tail of formal mathematics require Compositional Learning Behaviours (CLBs) -- the capacity to ground and recombine novel symbolic structures in context, beyond mere recombination of prelearned atoms. We propose \text…