Aletheia tackles FirstProof autonomously
arXiv: 2602.21201v1
发布: 2026-02-24
更新: 2026-02-24
AI 摘要
Aletheia基于Gemini 3在FirstProof数学挑战赛中自主解决了6/10的问题。
主要贡献
- 验证了Gemini 3 Deep Think在复杂数学问题上的推理能力
- 提出了一个自主解决数学问题的Agent Aletheia
- 公开了实验细节和评估方法,增强了透明度
方法论
Aletheia是一个基于Gemini 3 Deep Think的数学研究Agent,通过自主规划和推理解决FirstProof挑战中的数学问题。
原文摘要
We report the performance of Aletheia (Feng et al., 2026b), a mathematics research agent powered by Gemini 3 Deep Think, on the inaugural FirstProof challenge. Within the allowed timeframe of the challenge, Aletheia autonomously solved 6 problems (2, 5, 7, 8, 9, 10) out of 10 according to majority expert assessments; we note that experts were not unanimous on Problem 8 (only). For full transparency, we explain our interpretation of FirstProof and disclose details about our experiments as well as our evaluation. Raw prompts and outputs are available at https://github.com/google-deepmind/superhuman/tree/main/aletheia.