{"title":"实现自动功能方程证明:基准数据集和特定领域的上下文代理","authors":"Mahdi Buali, Robert Hoehndorf","doi":"arxiv-2407.14521","DOIUrl":null,"url":null,"abstract":"Automated Theorem Proving (ATP) faces challenges due to its complexity and\ncomputational demands. Recent work has explored using Large Language Models\n(LLMs) for ATP action selection, but these methods can be resource-intensive.\nThis study introduces FEAS, an agent that enhances the COPRA in-context\nlearning framework within Lean. FEAS refines prompt generation, response\nparsing, and incorporates domain-specific heuristics for functional equations.\nIt introduces FunEq, a curated dataset of functional equation problems with\nvarying difficulty. FEAS outperforms baselines on FunEq, particularly with the\nintegration of domain-specific heuristics. The results demonstrate FEAS's\neffectiveness in generating and formalizing high-level proof strategies into\nLean proofs, showcasing the potential of tailored approaches for specific ATP\nchallenges.","PeriodicalId":501033,"journal":{"name":"arXiv - CS - Symbolic Computation","volume":"94 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Towards Automated Functional Equation Proving: A Benchmark Dataset and A Domain-Specific In-Context Agent\",\"authors\":\"Mahdi Buali, Robert Hoehndorf\",\"doi\":\"arxiv-2407.14521\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Automated Theorem Proving (ATP) faces challenges due to its complexity and\\ncomputational demands. Recent work has explored using Large Language Models\\n(LLMs) for ATP action selection, but these methods can be resource-intensive.\\nThis study introduces FEAS, an agent that enhances the COPRA in-context\\nlearning framework within Lean. FEAS refines prompt generation, response\\nparsing, and incorporates domain-specific heuristics for functional equations.\\nIt introduces FunEq, a curated dataset of functional equation problems with\\nvarying difficulty. FEAS outperforms baselines on FunEq, particularly with the\\nintegration of domain-specific heuristics. The results demonstrate FEAS's\\neffectiveness in generating and formalizing high-level proof strategies into\\nLean proofs, showcasing the potential of tailored approaches for specific ATP\\nchallenges.\",\"PeriodicalId\":501033,\"journal\":{\"name\":\"arXiv - CS - Symbolic Computation\",\"volume\":\"94 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Symbolic Computation\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.14521\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Symbolic Computation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.14521","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
自动定理证明(ATP)因其复杂性和计算需求而面临挑战。最近的工作探索了使用大型语言模型(LLMs)进行 ATP 动作选择,但这些方法可能会耗费大量资源。本研究介绍了 FEAS,它是一种在 Lean 中增强 COPRA 上下文学习框架的代理。FEAS 改进了提示生成、响应解析,并纳入了针对特定领域的函数方程启发式。FEAS 引入了 FunEq,这是一个难度各异的函数方程问题数据集。FEAS 在 FunEq 上的表现优于基线,特别是在集成了特定领域启发式后。结果证明了 FEAS 在生成高层次证明策略并将其形式化为精益证明方面的有效性,展示了针对特定 ATP 挑战的定制方法的潜力。
Towards Automated Functional Equation Proving: A Benchmark Dataset and A Domain-Specific In-Context Agent
Automated Theorem Proving (ATP) faces challenges due to its complexity and
computational demands. Recent work has explored using Large Language Models
(LLMs) for ATP action selection, but these methods can be resource-intensive.
This study introduces FEAS, an agent that enhances the COPRA in-context
learning framework within Lean. FEAS refines prompt generation, response
parsing, and incorporates domain-specific heuristics for functional equations.
It introduces FunEq, a curated dataset of functional equation problems with
varying difficulty. FEAS outperforms baselines on FunEq, particularly with the
integration of domain-specific heuristics. The results demonstrate FEAS's
effectiveness in generating and formalizing high-level proof strategies into
Lean proofs, showcasing the potential of tailored approaches for specific ATP
challenges.