{"slug": "maxproof", "title": "Maxproof", "summary": "Researchers have developed MaxProof, a population-level test-time scaling framework for mathematical proof that enables the MiniMax-M3 model to achieve 35 out of 42 on IMO 2025 and 36 out of 42 on USAMO 2026, surpassing the human gold-medal threshold on both competitions. The system integrates proof generation, verification, and critique-conditioned repair into a single model, then searches over candidate proofs through tournament selection to produce a final output. This marks the first time an AI system has exceeded the gold-medal standard on these elite mathematical olympiads.", "body_md": "# Computer Science > Machine Learning\n\n[Submitted on 11 Jun 2026]\n\n# Title:MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling\n\n[View PDF](/pdf/2606.13473)\n\n[HTML (experimental)](https://arxiv.org/html/2606.13473v1)\n\nAbstract:We present MaxProof, a population-level test-time scaling framework for competition-level mathematical proof in the MiniMax-M3 series. M3 first trains three proof-oriented capabilities -- proof generation, proof verification, and critique-conditioned proof repair -- using a defense-in-depth generative verifier engineered for low false-positive rate. These capabilities are merged into a single released M3 model. At test time, MaxProof treats the model as a generator, verifier, refiner, and ranker, searches over a population of candidate proofs, and returns one final proof through tournament selection. With MaxProof test-time scaling, the M3 model reaches 35/42 on IMO 2025 and 36/42 on USAMO 2026, exceeding the human gold-medal threshold on both.\n\n### Current browse context:\n\ncs.LG\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\nIArxiv Recommender\n\n*(*[What is IArxiv?](https://iarxiv.org/about))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/maxproof", "canonical_source": "https://arxiv.org/abs/2606.13473", "published_at": "2026-06-12 12:00:02+00:00", "updated_at": "2026-06-12 12:42:58.387096+00:00", "lang": "en", "topics": ["machine-learning", "large-language-models", "artificial-intelligence", "ai-research"], "entities": ["MaxProof", "MiniMax-M3", "IMO", "USAMO"], "alternates": {"html": "https://wpnews.pro/news/maxproof", "markdown": "https://wpnews.pro/news/maxproof.md", "text": "https://wpnews.pro/news/maxproof.txt", "jsonld": "https://wpnews.pro/news/maxproof.jsonld"}}