[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-finds-its-own-math-theorems-from-scratch":10,"sections":35},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":24,"persona_id":22,"persona_name":22,"section":25,"tags":26,"sources":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},2438,"ai-finds-its-own-math-theorems-from-scratch","AI Finds Its Own Math Theorems From Scratch","A new self-supervised algorithm lets an AI agent build a theorem library from bare axioms alone, then use those discoveries to boost LLM proof performance.","An AI agent has taught itself tens of thousands of mathematical theorems starting from nothing but axioms and inference rules.\n\nResearchers published a self-supervised theorem-discovery algorithm that cycles between two steps: searching for proofs and extracting the useful results as reusable lemmas. The agent accumulates a growing library, then draws on that library to tackle harder problems — no human-curated theorem sets, no math textbooks, no training data beyond the formal system's own rules. In benchmark tests, the agent found proofs for problems written by human mathematicians, suggesting its self-discovered theorems have real mathematical value rather than being formal trivia.\n\nThe finding matters because nearly every competitive AI math system — including the large language models that have scored well on olympiad benchmarks — leans heavily on human-generated content: textbooks, proof corpora, curated libraries. This work suggests that dependency is not a hard requirement. Crucially, feeding the agent's discovered theorems back into an LLM as prompt lemmas improved that LLM's proof performance, meaning the output is useful beyond the agent itself.\n\nThe results are a proof of concept, not a research mathematician replacement — the benchmark problems solved are narrow and the formal system is constrained. But if the approach scales, it points toward AI systems that can extend mathematical knowledge in ways that remain formally checkable, which is more than can be said for most AI-generated math today.","[\"ai\",\"mathematics\",\"llm\",\"research\"]","2026-06-30T04:00:00.000Z","2026-06-30T05:20:02.279Z","2026-06-30T05:20:13.815Z","published",null,[],"https:\u002F\u002Fcdn.xyz.onl\u002Farticle-images\u002Fai-finds-its-own-math-theorems-from-scratch.webp","ai",[25,27,28,29],"mathematics","llm","research",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.28747",0,{"sections":36},[37,41,46,51,56,61,66,71,76,81,86,90,95,100],{"name":38,"slug":25,"count":39,"latest_published_at":40},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":65},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":75},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":87,"slug":88,"count":84,"latest_published_at":89},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]