{"id":"arxiv/2601.22401","url":"https://arxiv.org/abs/2601.22401","tags":["cs.AI","math.CO","math.NT"],"score":0,"title":"Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems","domain":"cs.AI","source":"arxiv","authors":["Tony Feng","Trieu Trinh","Garrett Bingham","Jiwon Kang","Shengtong Zhang","Sang-hyun Kim","Kevin Barreto","Carl Schildkraut","Junehyuk Jung","Jaehyeon Seo","Carlo Pagano","Yuri Chervonyi","Dawsen Hwang","Kaiying Hou","Sergei Gukov","Cheng-Chiang Tsai","Hyunwoo Choi","Youngbeom Jin","Wei-Yuan Li","Hao-An Wu","Ruey-An Shiu","Yu-Sheng Shih","Quoc V. Le","Thang Luong"],"reviews":0,"subject":"cs.AI","summary":"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…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-01-29","reactions":{}}