Paper
Aletheia tackles FirstProof autonomously
Authors
Tony Feng, Junehyuk Jung, Sang-hyun Kim, Carlo Pagano, Sergei Gukov, Chiang-Chiang Tsai, David Woodruff, Adel Javanmard, Aryan Mokhtari, Dawsen Hwang, Yuri Chervonyi, Jonathan N. Lee, Garrett Bingham, Trieu H. Trinh, Vahab Mirrokni, Quoc V. Le, Thang Luong
Abstract
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.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2602.21201v1</id>\n <title>Aletheia tackles FirstProof autonomously</title>\n <updated>2026-02-24T18:56:10Z</updated>\n <link href='https://arxiv.org/abs/2602.21201v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2602.21201v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>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.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n <published>2026-02-24T18:56:10Z</published>\n <arxiv:comment>34 pages. Project page: https://github.com/google-deepmind/superhuman/tree/main/aletheia</arxiv:comment>\n <arxiv:primary_category term='cs.AI'/>\n <author>\n <name>Tony Feng</name>\n </author>\n <author>\n <name>Junehyuk Jung</name>\n </author>\n <author>\n <name>Sang-hyun Kim</name>\n </author>\n <author>\n <name>Carlo Pagano</name>\n </author>\n <author>\n <name>Sergei Gukov</name>\n </author>\n <author>\n <name>Chiang-Chiang Tsai</name>\n </author>\n <author>\n <name>David Woodruff</name>\n </author>\n <author>\n <name>Adel Javanmard</name>\n </author>\n <author>\n <name>Aryan Mokhtari</name>\n </author>\n <author>\n <name>Dawsen Hwang</name>\n </author>\n <author>\n <name>Yuri Chervonyi</name>\n </author>\n <author>\n <name>Jonathan N. Lee</name>\n </author>\n <author>\n <name>Garrett Bingham</name>\n </author>\n <author>\n <name>Trieu H. Trinh</name>\n </author>\n <author>\n <name>Vahab Mirrokni</name>\n </author>\n <author>\n <name>Quoc V. Le</name>\n </author>\n <author>\n <name>Thang Luong</name>\n </author>\n </entry>"
}