Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456 Paper page - Dr. Zero: Self-Evolving Search Agents without Training Data
Please give a thumbs up to this comment if you found it helpful!
\n
If you want recommendations for any Paper on Hugging Face checkout this Space
\n
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend
\n","updatedAt":"2026-01-15T01:38:35.770Z","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":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7572200298309326},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}},{"id":"696b8c02ca5779cca92d1404","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false},"createdAt":"2026-01-17T13:17:54.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"arXivlens breakdown of this paper 👉 https://arxivlens.com/PaperView/Details/dr-zero-self-evolving-search-agents-without-training-data-1564-a632b389\n\n- Executive Summary\n- Detailed Breakdown\n- Practical Applications","html":"
\n","updatedAt":"2026-01-17T13:17:54.575Z","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7175832390785217},"editors":["avahal"],"editorAvatarUrls":["/avatars/743a009681d5d554c27e04300db9f267.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2601.07055","authors":[{"_id":"6965c753fc8c4ecc02c7f9c6","user":{"_id":"644c4eff9a7c19cfaf0c8f4a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/2gHtfhPFjSsnwxVtBnxe2.jpeg","isPro":false,"fullname":"Zhenrui Yue","user":"yueeeeeeee2837","type":"user"},"name":"Zhenrui Yue","status":"claimed_verified","statusLastChangedAt":"2026-01-14T09:53:09.699Z","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9c7","name":"Kartikeya Upasani","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9c8","name":"Xianjun Yang","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9c9","name":"Suyu Ge","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9ca","name":"Shaoliang Nie","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9cb","name":"Yuning Mao","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9cc","name":"Zhe Liu","hidden":false},{"_id":"6965c753fc8c4ecc02c7f9cd","name":"Dong Wang","hidden":false}],"publishedAt":"2026-01-11T20:27:55.000Z","submittedOnDailyAt":"2026-01-13T01:47:35.465Z","title":"Dr. Zero: Self-Evolving Search Agents without Training Data","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},"summary":"As high-quality data becomes increasingly difficult to obtain, data-free self-evolution has emerged as a promising paradigm. This approach allows large language models (LLMs) to autonomously generate and solve complex problems, thereby improving their reasoning capabilities. However, multi-turn search agents struggle in data-free self-evolution due to the limited question diversity and the substantial compute required for multi-step reasoning and tool using. In this work, we introduce Dr. Zero, a framework enabling search agents to effectively self-evolve without any training data. In particular, we design a self-evolution feedback loop where a proposer generates diverse questions to train a solver initialized from the same base model. As the solver evolves, it incentivizes the proposer to produce increasingly difficult yet solvable tasks, thus establishing an automated curriculum to refine both agents. To enhance training efficiency, we also introduce hop-grouped relative policy optimization (HRPO). This method clusters structurally similar questions to construct group-level baselines, effectively minimizing the sampling overhead in evaluating each query's individual difficulty and solvability. Consequently, HRPO significantly reduces the compute requirements for solver training without compromising performance or stability. Extensive experiment results demonstrate that the data-free Dr. Zero matches or surpasses fully supervised search agents, proving that complex reasoning and search capabilities can emerge solely through self-evolution.","upvotes":21,"discussionId":"6965c753fc8c4ecc02c7f9ce","githubRepo":"https://github.com/facebookresearch/drzero","githubRepoAddedBy":"user","ai_summary":"A data-free self-evolution framework enables large language models to autonomously improve reasoning capabilities through iterative question generation and solving, achieving performance comparable to supervised methods.","ai_keywords":["large language models","self-evolution","search agents","proposer","solver","automated curriculum","hop-grouped relative policy optimization","HRPO","multi-turn reasoning","tool using","question diversity","compute requirements","training efficiency","performance","stability"],"githubStars":463},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6270324ebecab9e2dcf245de","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6270324ebecab9e2dcf245de/cMbtWSasyNlYc9hvsEEzt.jpeg","isPro":false,"fullname":"Kye Gomez","user":"kye","type":"user"},{"_id":"644c4eff9a7c19cfaf0c8f4a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/2gHtfhPFjSsnwxVtBnxe2.jpeg","isPro":false,"fullname":"Zhenrui Yue","user":"yueeeeeeee2837","type":"user"},{"_id":"62f662bcc58915315c4eccea","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62f662bcc58915315c4eccea/zOAQLONfMP88zr70sxHK-.jpeg","isPro":true,"fullname":"Yilun Zhao","user":"yilunzhao","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":"63082bb7bc0a2a5ee2253523","avatarUrl":"/avatars/6cf8d12d16d15db1070fbea89b5b3967.svg","isPro":false,"fullname":"Kuo-Hsin Tu","user":"dapumptu","type":"user"},{"_id":"684d57f26e04c265777ead3f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/cuOj-bQqukSZreXgUJlfm.png","isPro":false,"fullname":"Joakim Lee","user":"Reinforcement4All","type":"user"},{"_id":"62ea79dd01ed9b0e8f61ccd3","avatarUrl":"/avatars/70af83e0e267be39fcd5f23b85e2dafa.svg","isPro":false,"fullname":"Chengsong Huang","user":"ChengsongHuang","type":"user"},{"_id":"655879337b098f9cb5dbdb16","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/655879337b098f9cb5dbdb16/j-NwJgRvi_2jw6JU0yFp4.jpeg","isPro":false,"fullname":"Ljubomir Josifovski","user":"ljupco","type":"user"},{"_id":"65c4063740d617a14238f3df","avatarUrl":"/avatars/726b1470e46ad71c9ec233f3f0f396ec.svg","isPro":false,"fullname":"Zikun Li","user":"zikun-li","type":"user"},{"_id":"64d4615cf8082bf19b916492","avatarUrl":"/avatars/8e1b59565ec5e4b31090cf1b911781b9.svg","isPro":false,"fullname":"wongyukim","user":"wongyukim","type":"user"},{"_id":"68e7b8b015bfb894fff9b3ba","avatarUrl":"/avatars/dc2a1e8722b52d6bf9f22edd6deea0b1.svg","isPro":true,"fullname":"J G","user":"jgar7","type":"user"},{"_id":"62c16d9aa20c0353395c0ef1","avatarUrl":"/avatars/d0868ac895df9087ee4e23ae6504faf3.svg","isPro":false,"fullname":"Gurumurthi V Ramanan","user":"GVR","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">
A data-free self-evolution framework enables large language models to autonomously improve reasoning capabilities through iterative question generation and solving, achieving performance comparable to supervised methods.
AI-generated summary
As high-quality data becomes increasingly difficult to obtain, data-free self-evolution has emerged as a promising paradigm. This approach allows large language models (LLMs) to autonomously generate and solve complex problems, thereby improving their reasoning capabilities. However, multi-turn search agents struggle in data-free self-evolution due to the limited question diversity and the substantial compute required for multi-step reasoning and tool using. In this work, we introduce Dr. Zero, a framework enabling search agents to effectively self-evolve without any training data. In particular, we design a self-evolution feedback loop where a proposer generates diverse questions to train a solver initialized from the same base model. As the solver evolves, it incentivizes the proposer to produce increasingly difficult yet solvable tasks, thus establishing an automated curriculum to refine both agents. To enhance training efficiency, we also introduce hop-grouped relative policy optimization (HRPO). This method clusters structurally similar questions to construct group-level baselines, effectively minimizing the sampling overhead in evaluating each query's individual difficulty and solvability. Consequently, HRPO significantly reduces the compute requirements for solver training without compromising performance or stability. Extensive experiment results demonstrate that the data-free Dr. Zero matches or surpasses fully supervised search agents, proving that complex reasoning and search capabilities can emerge solely through self-evolution.
Dr. Zero enables data-free self-evolving search agents through a self-evolution loop with HRPO, achieving strong multi-step reasoning while reducing compute.