Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.22401 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914307414949888 |
|---|---|
| author | Feng, Tony Trinh, Trieu Bingham, Garrett Kang, Jiwon Zhang, Shengtong Kim, Sang-hyun Barreto, Kevin Schildkraut, Carl Jung, Junehyuk Seo, Jaehyeon Pagano, Carlo Chervonyi, Yuri Hwang, Dawsen Hou, Kaiying Gukov, Sergei Tsai, Cheng-Chiang Choi, Hyunwoo Jin, Youngbeom Li, Wei-Yuan Wu, Hao-An Shiu, Ruey-An Shih, Yu-Sheng Le, Quoc V. Luong, Thang |
| author_facet | Feng, Tony Trinh, Trieu Bingham, Garrett Kang, Jiwon Zhang, Shengtong Kim, Sang-hyun Barreto, Kevin Schildkraut, Carl Jung, Junehyuk Seo, Jaehyeon Pagano, Carlo Chervonyi, Yuri Hwang, Dawsen Hou, Kaiying Gukov, Sergei Tsai, Cheng-Chiang Choi, Hyunwoo Jin, Youngbeom Li, Wei-Yuan Wu, Hao-An Shiu, Ruey-An Shih, Yu-Sheng Le, Quoc V. Luong, Thang |
| contents | We present a case study in semi-autonomous mathematics discovery, using Gemini to systematically evaluate 700 conjectures labeled 'Open' in Bloom's Erdős Problems database. We employ a hybrid methodology: AI-driven natural language verification to narrow the search space, followed by human expert evaluation to gauge correctness and novelty. We address 13 problems that were marked 'Open' in the database: 5 through seemingly novel autonomous solutions, and 8 through identification of previous solutions in the existing literature. Our findings suggest that the 'Open' status of the problems was through obscurity rather than difficulty. We also identify and discuss issues arising in applying AI to math conjectures at scale, highlighting the difficulty of literature identification and the risk of ''subconscious plagiarism'' by AI. We reflect on the takeaways from AI-assisted efforts on the Erdős Problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22401 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems Feng, Tony Trinh, Trieu Bingham, Garrett Kang, Jiwon Zhang, Shengtong Kim, Sang-hyun Barreto, Kevin Schildkraut, Carl Jung, Junehyuk Seo, Jaehyeon Pagano, Carlo Chervonyi, Yuri Hwang, Dawsen Hou, Kaiying Gukov, Sergei Tsai, Cheng-Chiang Choi, Hyunwoo Jin, Youngbeom Li, Wei-Yuan Wu, Hao-An Shiu, Ruey-An Shih, Yu-Sheng Le, Quoc V. Luong, Thang Artificial Intelligence Combinatorics Number Theory We present a case study in semi-autonomous mathematics discovery, using Gemini to systematically evaluate 700 conjectures labeled 'Open' in Bloom's Erdős Problems database. We employ a hybrid methodology: AI-driven natural language verification to narrow the search space, followed by human expert evaluation to gauge correctness and novelty. We address 13 problems that were marked 'Open' in the database: 5 through seemingly novel autonomous solutions, and 8 through identification of previous solutions in the existing literature. Our findings suggest that the 'Open' status of the problems was through obscurity rather than difficulty. We also identify and discuss issues arising in applying AI to math conjectures at scale, highlighting the difficulty of literature identification and the risk of ''subconscious plagiarism'' by AI. We reflect on the takeaways from AI-assisted efforts on the Erdős Problems. |
| title | Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems |
| topic | Artificial Intelligence Combinatorics Number Theory |
| url | https://arxiv.org/abs/2601.22401 |