https://github.com/SakanaAI/AI-Scientist-v2

\n","updatedAt":"2025-04-15T10:21:00.362Z","author":{"_id":"63142272e29fb2e86d63db0a","avatarUrl":"/avatars/5726b5c40cea190a6cd9907fa596b66b.svg","fullname":"Yutaro","name":"yyamada","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8922260999679565},"editors":["yyamada"],"editorAvatarUrls":["/avatars/5726b5c40cea190a6cd9907fa596b66b.svg"],"reactions":[],"isReport":false}},{"id":"6800bd36e32bd8aecd0e60ae","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264},"createdAt":"2025-04-17T08:35:02.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [Evaluating Sakana's AI Scientist for Autonomous Research: Wishful Thinking or an Emerging Reality Towards 'Artificial Research Intelligence' (ARI)?](https://huggingface.co/papers/2502.14297) (2025)\n* [Towards an AI co-scientist](https://huggingface.co/papers/2502.18864) (2025)\n* [CodeScientist: End-to-End Semi-Automated Scientific Discovery with Code-based Experimentation](https://huggingface.co/papers/2503.22708) (2025)\n* [Agentic AI for Scientific Discovery: A Survey of Progress, Challenges, and Future Directions](https://huggingface.co/papers/2503.08979) (2025)\n* [The AI Cosmologist I: An Agentic System for Automated Data Analysis](https://huggingface.co/papers/2504.03424) (2025)\n* [Scaling Laws in Scientific Discovery with AI and Robot Scientists](https://huggingface.co/papers/2503.22444) (2025)\n* [SciSciGPT: Advancing Human-AI Collaboration in the Science of Science](https://huggingface.co/papers/2504.05559) (2025)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2025-04-17T08:35:02.008Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7268210053443909},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}},{"id":"683b557ac15aff506f41575c","author":{"_id":"64b655948f25ae1dd3afa552","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/JlLx6zb-QcMSyfOhz_QmZ.png","fullname":"Kian Kyars","name":"kyars","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false},"createdAt":"2025-05-31T19:16:10.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Exciting.\n","html":"

Exciting.

\n","updatedAt":"2025-05-31T19:16:10.879Z","author":{"_id":"64b655948f25ae1dd3afa552","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/JlLx6zb-QcMSyfOhz_QmZ.png","fullname":"Kian Kyars","name":"kyars","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9418047666549683},"editors":["kyars"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/JlLx6zb-QcMSyfOhz_QmZ.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2504.08066","authors":[{"_id":"67fd9abb4f2e48847878c9e1","user":{"_id":"63142272e29fb2e86d63db0a","avatarUrl":"/avatars/5726b5c40cea190a6cd9907fa596b66b.svg","isPro":false,"fullname":"Yutaro","user":"yyamada","type":"user"},"name":"Yutaro Yamada","status":"extracted_confirmed","statusLastChangedAt":"2025-04-15T11:14:40.840Z","hidden":false},{"_id":"67fd9abb4f2e48847878c9e2","name":"Robert Tjarko Lange","hidden":false},{"_id":"67fd9abb4f2e48847878c9e3","user":{"_id":"6311ec6ac7722fdac9a93f92","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670192227454-6311ec6ac7722fdac9a93f92.jpeg","isPro":false,"fullname":"Cong Lu","user":"conglu","type":"user"},"name":"Cong Lu","status":"claimed_verified","statusLastChangedAt":"2025-04-15T07:55:14.712Z","hidden":false},{"_id":"67fd9abb4f2e48847878c9e4","user":{"_id":"640530fd0ab5e22719fb199a","avatarUrl":"/avatars/6b023f496a2f8cb5f37ec617d84179b7.svg","isPro":false,"fullname":"Shengran HU","user":"Shengran","type":"user"},"name":"Shengran Hu","status":"claimed_verified","statusLastChangedAt":"2025-06-02T07:51:03.607Z","hidden":false},{"_id":"67fd9abb4f2e48847878c9e5","name":"Chris Lu","hidden":false},{"_id":"67fd9abb4f2e48847878c9e6","name":"Jakob Foerster","hidden":false},{"_id":"67fd9abb4f2e48847878c9e7","name":"Jeff Clune","hidden":false},{"_id":"67fd9abb4f2e48847878c9e8","name":"David Ha","hidden":false}],"publishedAt":"2025-04-10T18:44:41.000Z","submittedOnDailyAt":"2025-04-15T08:51:00.354Z","title":"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via\n Agentic Tree Search","submittedOnDailyBy":{"_id":"63142272e29fb2e86d63db0a","avatarUrl":"/avatars/5726b5c40cea190a6cd9907fa596b66b.svg","isPro":false,"fullname":"Yutaro","user":"yyamada","type":"user"},"summary":"AI is increasingly playing a pivotal role in transforming how scientific\ndiscoveries are made. We introduce The AI Scientist-v2, an end-to-end agentic\nsystem capable of producing the first entirely AI generated\npeer-review-accepted workshop paper. This system iteratively formulates\nscientific hypotheses, designs and executes experiments, analyzes and\nvisualizes data, and autonomously authors scientific manuscripts. Compared to\nits predecessor (v1, Lu et al., 2024 arXiv:2408.06292), The AI Scientist-v2\neliminates the reliance on human-authored code templates, generalizes\neffectively across diverse machine learning domains, and leverages a novel\nprogressive agentic tree-search methodology managed by a dedicated experiment\nmanager agent. Additionally, we enhance the AI reviewer component by\nintegrating a Vision-Language Model (VLM) feedback loop for iterative\nrefinement of content and aesthetics of the figures. We evaluated The AI\nScientist-v2 by submitting three fully autonomous manuscripts to a\npeer-reviewed ICLR workshop. Notably, one manuscript achieved high enough\nscores to exceed the average human acceptance threshold, marking the first\ninstance of a fully AI-generated paper successfully navigating a peer review.\nThis accomplishment highlights the growing capability of AI in conducting all\naspects of scientific research. We anticipate that further advancements in\nautonomous scientific discovery technologies will profoundly impact human\nknowledge generation, enabling unprecedented scalability in research\nproductivity and significantly accelerating scientific breakthroughs, greatly\nbenefiting society at large. We have open-sourced the code at\nhttps://github.com/SakanaAI/AI-Scientist-v2 to foster the future development of\nthis transformative technology. We also discuss the role of AI in science,\nincluding AI safety.","upvotes":14,"discussionId":"67fd9abd4f2e48847878ca5b","ai_summary":"The AI Scientist-v2 autonomously proposes hypotheses, performs experiments, analyzes data, and writes peer-reviewed scientific papers, marking the first fully AI-generated paper accepted by a conference.","ai_keywords":["agentic system","scientific hypotheses","experiment design","data analysis","autonomous authoring","machine learning domains","progressive agentic tree-search","experiment manager agent","Vision-Language Model (VLM)","peer-reviewed ICLR workshop","AI safety"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6311ec6ac7722fdac9a93f92","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670192227454-6311ec6ac7722fdac9a93f92.jpeg","isPro":false,"fullname":"Cong Lu","user":"conglu","type":"user"},{"_id":"648eb1eb59c4e5c87dc116e0","avatarUrl":"/avatars/c636cea39c2c0937f01398c94ead5dad.svg","isPro":false,"fullname":"fdsqefsgergd","user":"T-representer","type":"user"},{"_id":"63b6f2e752c02ae8acbaa4d8","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1672934038280-noauth.jpeg","isPro":false,"fullname":"Habibullah Akbar","user":"ChavyvAkvar","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"654b3930f8ebcec54532a1da","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/1BDZEbwzCccR8yNxzjKbf.png","isPro":false,"fullname":"Schneider Jérémy","user":"Picrik","type":"user"},{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},{"_id":"6270324ebecab9e2dcf245de","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6270324ebecab9e2dcf245de/cMbtWSasyNlYc9hvsEEzt.jpeg","isPro":false,"fullname":"Kye Gomez","user":"kye","type":"user"},{"_id":"6751ae913f59c62f77583757","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/_dCyD7vTFPlFETR97WyQj.png","isPro":false,"fullname":"Ivy Zhang","user":"Ivy1997","type":"user"},{"_id":"6302d6ed2db53f7d9f3a9284","avatarUrl":"/avatars/0da00202946ef080adffee9412efb329.svg","isPro":false,"fullname":"Agusti","user":"agust","type":"user"},{"_id":"64d89d015900b6d11116dab0","avatarUrl":"/avatars/fc80ff8df515b9cb4de48ec894539ed1.svg","isPro":false,"fullname":"Zhiyuan Ning","user":"nzynzy","type":"user"},{"_id":"64b655948f25ae1dd3afa552","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/JlLx6zb-QcMSyfOhz_QmZ.png","isPro":false,"fullname":"Kian Kyars","user":"kyars","type":"user"},{"_id":"64daeadd70446182be66f355","avatarUrl":"/avatars/7595098f12b65866087a146b5c1a3133.svg","isPro":false,"fullname":"Ryo Sakai","user":"rsakai","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2504.08066

The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search

Published on Apr 10

· Submitted by

Yutaro on Apr 15

Upvote

Authors:

Yutaro Yamada ,

Cong Lu ,

Shengran Hu ,

Abstract

The AI Scientist-v2 autonomously proposes hypotheses, performs experiments, analyzes data, and writes peer-reviewed scientific papers, marking the first fully AI-generated paper accepted by a conference.

AI-generated summary

AI is increasingly playing a pivotal role in transforming how scientific discoveries are made. We introduce The AI Scientist-v2, an end-to-end agentic system capable of producing the first entirely AI generated peer-review-accepted workshop paper. This system iteratively formulates scientific hypotheses, designs and executes experiments, analyzes and visualizes data, and autonomously authors scientific manuscripts. Compared to its predecessor (v1, Lu et al., 2024 arXiv:2408.06292), The AI Scientist-v2 eliminates the reliance on human-authored code templates, generalizes effectively across diverse machine learning domains, and leverages a novel progressive agentic tree-search methodology managed by a dedicated experiment manager agent. Additionally, we enhance the AI reviewer component by integrating a Vision-Language Model (VLM) feedback loop for iterative refinement of content and aesthetics of the figures. We evaluated The AI Scientist-v2 by submitting three fully autonomous manuscripts to a peer-reviewed ICLR workshop. Notably, one manuscript achieved high enough scores to exceed the average human acceptance threshold, marking the first instance of a fully AI-generated paper successfully navigating a peer review. This accomplishment highlights the growing capability of AI in conducting all aspects of scientific research. We anticipate that further advancements in autonomous scientific discovery technologies will profoundly impact human knowledge generation, enabling unprecedented scalability in research productivity and significantly accelerating scientific breakthroughs, greatly benefiting society at large. We have open-sourced the code at https://github.com/SakanaAI/AI-Scientist-v2 to foster the future development of this transformative technology. We also discuss the role of AI in science, including AI safety.

View arXiv page View PDF Add to collection

Community

yyamada

Paper author Paper submitter Apr 15

The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search

Fully autonomous scientific research systems are becoming increasingly capable, with AI playing a pivotal role in transforming how scientific discoveries are made. We are excited to introduce The AI Scientist-v2, a generalized end-to-end agentic system that has generated the first workshop paper written entirely by AI and accepted through peer review.

This system autonomously generates hypotheses, runs experiments, analyzes data, and writes scientific manuscripts. Unlike its predecessor, the AI Scientist-v2 removes reliance on human-authored templates, generalizes across ML domains, and employs a progressive agentic tree-search guided by an experiment manager agent. Code is available at https://github.com/SakanaAI/AI-Scientist-v2