Hugging Face Daily Papers · September 15, 2026 · 4 min read

Discovery Foundation Models: Toward Open-Ended Discovery Intelligence

Mirrored from Hugging Face Daily Papers for archival readability. Support the source by reading on the original site.

Like Read original ↗

<a href=\"https://github.com/Gen-Verse/DFM-Plans\" rel=\"nofollow\">https://github.com/Gen-Verse/DFM-Plans</a></p>\n","updatedAt":"2026-09-15T02:29:21.662Z","author":{"_id":"64fde4e252e82dd432b74ce9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64fde4e252e82dd432b74ce9/-CQZbBP7FsPPyawYrsi4z.jpeg","fullname":"Ling Yang","name":"Lingaaaaaaa","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":16,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6921064257621765},"editors":["Lingaaaaaaa"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/64fde4e252e82dd432b74ce9/-CQZbBP7FsPPyawYrsi4z.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2609.15973","authors":[{"_id":"6aa8ad375dd4cb9b4cc0250a","name":"Ling Yang","hidden":false},{"_id":"6aa8ad375dd4cb9b4cc0250b","name":"Zhenfei Yin","hidden":false},{"_id":"6aa8ad375dd4cb9b4cc0250c","name":"Yingcheng Wu","hidden":false}],"publishedAt":"2026-09-14T00:00:00.000Z","submittedOnDailyAt":"2026-09-15T00:00:00.000Z","title":"Discovery Foundation Models: Toward Open-Ended Discovery Intelligence","submittedOnDailyBy":{"_id":"64fde4e252e82dd432b74ce9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64fde4e252e82dd432b74ce9/-CQZbBP7FsPPyawYrsi4z.jpeg","isPro":false,"fullname":"Ling Yang","user":"Lingaaaaaaa","type":"user","name":"Lingaaaaaaa"},"summary":"Foundation models have progressed from learning and reasoning over existing knowledge, to increasingly learning through action, tool use, and outcome feedback. We argue that the next frontier is a further transition: from solving and acting within problems specified by humans to participating in the process by which new problems, representations, explanations, and knowledge are created. We refer to this capability as Discovery Intelligence. We formulate Discovery Foundation Models (DFMs) as general-purpose model systems for open-ended discovery. A DFM operates over a revisable research state and supports seven coupled capabilities spanning problem discovery, formulation, representation construction, hypothesis formation, intervention, evidence-grounded revision, and continual discovery improvement. We instantiate this framework with Zetema, which couples explicit research-state dynamics, verification and experimental gating, external grounding, and cross-task Discovery Skill evolution. We further ground the framework with GALILEO, a real therapeutic-discovery system in which Dry-Lab reasoning, robotic and hands-on Wet-Lab experimentation, external biological evidence, and iterative hypothesis and design revision form a closed physical discovery loop. We then formulate a unified approach to capability formation and process-centered evaluation, enabling discovery behavior to be trained, improved, and measured beyond final-answer performance. Together, these components establish discovery as a learnable, executable, and evaluable capability of foundation-model systems. We view this shift as a broader progression in intelligence scaling: from learning over existing knowledge, to learning from action outcomes, and ultimately to participating in the construction, testing, and revision of the structures through which new knowledge is discovered. Code: https://github.com/Gen-Verse/DFM-Plans","upvotes":18,"discussionId":"6aa8ad385dd4cb9b4cc0250d","projectPage":"https://phai-labs.com/collaborate/","githubRepo":"https://github.com/Gen-Verse/DFM-Plans","githubRepoAddedBy":"user","ai_summary":"Discovery Foundation Models enable open-ended scientific discovery through iterative problem formulation, hypothesis testing, and evidence-based revision across dry and wet lab settings.","ai_keywords":["Discovery Intelligence","Discovery Foundation Models","research state","problem discovery","representation construction","hypothesis formation","evidence-grounded revision","continual discovery improvement","Zetema","Discovery Skill","GALILEO","Dry-Lab reasoning","Wet-Lab experimentation","closed physical discovery loop","process-centered evaluation"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":2},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64fde4e252e82dd432b74ce9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64fde4e252e82dd432b74ce9/-CQZbBP7FsPPyawYrsi4z.jpeg","isPro":false,"fullname":"Ling Yang","user":"Lingaaaaaaa","type":"user"},{"_id":"641b26ed1911d3be6743e8d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/641b26ed1911d3be6743e8d0/oybjQjcEgiMBC-qGKCQVR.png","isPro":false,"fullname":"余昭辰","user":"BitStarWalkin","type":"user"},{"_id":"64e314ad24809d7fa0f20fbc","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/bHE0w_hjDFvU-Aul0_E7g.jpeg","isPro":false,"fullname":"Zhenfei Yin","user":"JeremyYin","type":"user"},{"_id":"6a8e4d98a6cfbc25447acd4b","avatarUrl":"/avatars/c6098454bad143112ef467791e0c6795.svg","isPro":false,"fullname":"zcy","user":"trafalgaryu","type":"user"},{"_id":"6a8e4ddef8b010c865604e6f","avatarUrl":"/avatars/e65b79944fa205555d8ceaeadcc6d24b.svg","isPro":false,"fullname":"sada","user":"fakersda","type":"user"},{"_id":"698196075cc02c68e5555ea5","avatarUrl":"/avatars/b36383b891f37275386b18f19d0a6ef3.svg","isPro":false,"fullname":"liu","user":"2linda","type":"user"},{"_id":"6981942081c01373225279f5","avatarUrl":"/avatars/b84b75986a3889ccb241b8a25cb09df8.svg","isPro":false,"fullname":"ma","user":"sherryma23","type":"user"},{"_id":"6662a59cf8d1fcc749cbc5de","avatarUrl":"/avatars/0e965b6b996c154b8d39106c0cc5178d.svg","isPro":false,"fullname":"liu","user":"miao99","type":"user"},{"_id":"6981928e9dcf301b73fe1fcb","avatarUrl":"/avatars/42d73f5498c65a4ebe67c168abafa218.svg","isPro":false,"fullname":"yang","user":"martinyang1","type":"user"},{"_id":"69817bd8a819c22bf570fae4","avatarUrl":"/avatars/00642747bfb2bcf443165ae7a4175c0c.svg","isPro":false,"fullname":"Yang","user":"TonyYang1","type":"user"},{"_id":"6662a23c2f86097c6d828b96","avatarUrl":"/avatars/2aa31ab30874257529861f2e4024acc2.svg","isPro":false,"fullname":"liu","user":"miao6","type":"user"},{"_id":"6aa8b0d172be0b16b8bc1ed7","avatarUrl":"/avatars/46bfb3db46bc8a08aa2f912d72011621.svg","isPro":false,"fullname":"asdz","user":"hle88123","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2609/2609.15973.md","query":{}}">

Papers

arxiv:2609.15973

Discovery Foundation Models: Toward Open-Ended Discovery Intelligence

Published on Sep 14

· Submitted by

Ling Yang on Sep 15

Upvote

Authors:

Abstract

Discovery Foundation Models enable open-ended scientific discovery through iterative problem formulation, hypothesis testing, and evidence-based revision across dry and wet lab settings.

Generated by thinkingmachines/Inkling-Small

Foundation models have progressed from learning and reasoning over existing knowledge, to increasingly learning through action, tool use, and outcome feedback. We argue that the next frontier is a further transition: from solving and acting within problems specified by humans to participating in the process by which new problems, representations, explanations, and knowledge are created. We refer to this capability as Discovery Intelligence. We formulate Discovery Foundation Models (DFMs) as general-purpose model systems for open-ended discovery. A DFM operates over a revisable research state and supports seven coupled capabilities spanning problem discovery, formulation, representation construction, hypothesis formation, intervention, evidence-grounded revision, and continual discovery improvement. We instantiate this framework with Zetema, which couples explicit research-state dynamics, verification and experimental gating, external grounding, and cross-task Discovery Skill evolution. We further ground the framework with GALILEO, a real therapeutic-discovery system in which Dry-Lab reasoning, robotic and hands-on Wet-Lab experimentation, external biological evidence, and iterative hypothesis and design revision form a closed physical discovery loop. We then formulate a unified approach to capability formation and process-centered evaluation, enabling discovery behavior to be trained, improved, and measured beyond final-answer performance. Together, these components establish discovery as a learnable, executable, and evaluable capability of foundation-model systems. We view this shift as a broader progression in intelligence scaling: from learning over existing knowledge, to learning from action outcomes, and ultimately to participating in the construction, testing, and revision of the structures through which new knowledge is discovered. Code: https://github.com/Gen-Verse/DFM-Plans