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 2602.10177cs.LGcs.AIcs.CLcs.CY
收藏

迈向自主数学研究

Towards Autonomous Mathematics Research

UC Berkeley(伯克利大学);Brown University(布朗大学);Yonsei University(延世大学);Concordia University(Concordia 大学);Korea Institute for Advanced Study(韩国高级研究院);UC San Diego(圣地亚哥大学);Caltech(加州理工学院)
Tony Feng, Trieu H. Trinh, Garrett Bingham, Dawsen Hwang, Yuri Chervonyi, Junehyuk Jung, Joonkyung Lee, Carlo Pagano, Sang-hyun Kim, Federico Pasqualotto, Serge… 展开作者

Tony Feng, Trieu H. Trinh, Garrett Bingham, Dawsen Hwang, Yuri Chervonyi, Junehyuk Jung, Joonkyung Lee, Carlo Pagano, Sang-hyun Kim, Federico Pasqualotto, Sergei Gukov, Jonathan N. Lee, Junsu Kim, Kaiying Hou, Golnaz Ghiasi, Yi Tay, YaGuang Li, Chenkai Kuang, Yuan Liu, Hanzhao Lin, Evan Zheran Liu, Nigamaa Nayakanti, Xiaomeng Yang, Heng-Tze Cheng, Demis Hassabis, Koray Kavukcuoglu, Quoc V. Le, Thang Luong

AI总结:

Aletheia是一个能够自主生成、验证和修订数学解决方案的代理,展示了AI在数学研究中的应用,包括生成研究论文和解决开放问题。

摘要:

近年来,基础模型的进步使得推理系统能够达到国际数学奥林匹克竞赛的金牌标准。然而,从竞赛级别的问题解决过渡到专业研究,需要在庞大的文献中导航并构建长周期的证明。在本工作中,我们引入了Aletheia,一个数学研究代理,能够以自然语言迭代生成、验证和修订解决方案。具体而言,Aletheia由先进的Gemini Deep Think版本驱动,用于解决具有挑战性的推理问题,一种新的推理时间扩展定律,使其能够超越奥林匹克级别问题,以及密集的工具使用来应对数学研究的复杂性。我们展示了Aletheia从奥林匹克问题到博士级别练习的能力,并且特别通过几个不同的AI辅助数学研究里程碑:(a)由AI生成的研究论文(Feng26),在没有人类干预计算某些算术几何结构常数(称为特征权重)的情况下;(b)展示人类与AI合作证明相互作用粒子系统界限的研究论文(LeeSeo26)称为独立集;以及(c)对Bloom的Erdos猜想数据库中700个开放问题的广泛半自动评估(Feng等人,2026a),包括对四个开放问题的自动解决方案。为了帮助公众更好地理解AI与数学发展的相关性,我们建议量化AI辅助结果的标准自主性和新颖性水平,并提出一个新颖的人机交互卡概念以提高透明度。我们最后反思了数学中的人机协作,并分享了所有提示以及模型输出在https://github.com/google-deepmind/superhuman/tree/main/aletheia。

英文摘要:

Recent advances in foundational models have yielded reasoning systems capable of achieving a gold-medal standard at the International Mathematical Olympiad. The transition from competition-level problem-solving to professional research, however, requires navigating vast literature and constructing long-horizon proofs. In this work, we introduce Aletheia, a math research agent that iteratively generates, verifies, and revises solutions end-to-end in natural language. Specifically, Aletheia is powered by an advanced version of Gemini Deep Think for challenging reasoning problems, a novel inference-time scaling law that extends beyond Olympiad-level problems, and intensive tool use to navigate the complexities of mathematical research. We demonstrate the capability of Aletheia from Olympiad problems to PhD-level exercises and most notably, through several distinct milestones in AI-assisted mathematics research: (a) a research paper (Feng26) generated by AI without any human intervention in calculating certain structure constants in arithmetic geometry called eigenweights; (b) a research paper (LeeSeo26) demonstrating human-AI collaboration in proving bounds on systems of interacting particles called independent sets; and (c) an extensive semi-autonomous evaluation (Feng et al., 2026a) of 700 open problems on Bloom's Erdos Conjectures database, including autonomous solutions to four open questions. In order to help the public better understand the developments pertaining to AI and mathematics, we suggest quantifying standard levels of autonomy and novelty of AI-assisted results, as well as propose a novel concept of human-AI interaction cards for transparency. We conclude with reflections on human-AI collaboration in mathematics and share all prompts as well as model outputs at https://github.com/google-deepmind/superhuman/tree/main/aletheia.

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。

↑