摘要
本文未声称证明黎曼假设，而是探讨可验证的 AI 辅助推理系统能否在明确识别剩余数学障碍的同时，产生可靠且经形式化检查的部分进展。研究应用具备递归自我改进能力的可验证增长物理 Transformer（VGPT-RSI）于两项任务：构建并验证参数化安全下曲线上的有限 RH 边界证书，以及启动形式化拉加里亚斯路径证书。系统利用区间算术和 Coq 进行严格审计，最终精准定位了未解决的数学瓶颈，展示了其在组织证明依赖和避免过度宣称方面的能力。
AI 推荐理由
论文核心研究可验证 AI 推理系统在数学证明中的应用与形式化检查机制。
研究机构
Ministry of Education Key Laboratory of Contemporary Anthropology, School of Life Sciences, Fudan University, Shanghai, China
Department of Immunology and Molecular Microbiology, School of Medicine, Texas Tech University Health Science Center, Lubbock, TX 79430, USA
Biostatistics and Bioinformatics Shared Resource, Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, Miami, FL 33136
School of Life Sciences, Fudan University, Shanghai 200438, China
Department of Biostatistics and Data Science, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX 77030, USA
Society of Artificial Intelligence Research, Houston, TX 77030, USA
论文信息