We release QUEST, a family of open models (ranging from 2B to 35B) that serve as general-purpose deep research agents designed to handle a wide range of long-horizon search tasks, with strong capabilities in fact seeking, citation grounding, and report synthesis.</p>\n","updatedAt":"2026-05-26T05:18:03.369Z","author":{"_id":"62d65139667051e0a29bffe7","avatarUrl":"/avatars/0252aa2bcd4cf1c8e4b87e5f164b6da5.svg","fullname":"Jian Xie","name":"hsaest","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":13,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9359260201454163},"editors":["hsaest"],"editorAvatarUrls":["/avatars/0252aa2bcd4cf1c8e4b87e5f164b6da5.svg"],"reactions":[{"reaction":"🚀","users":["drogozhang","nnnyt","Lzy01241010"],"count":3},{"reaction":"🔥","users":["drogozhang","nnnyt","Lzy01241010"],"count":3},{"reaction":"🤗","users":["drogozhang","nnnyt","Lzy01241010"],"count":3}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.24218","authors":[{"_id":"6a152c7cb57a1823d5708b70","user":{"_id":"62d65139667051e0a29bffe7","avatarUrl":"/avatars/0252aa2bcd4cf1c8e4b87e5f164b6da5.svg","isPro":false,"fullname":"Jian Xie","user":"hsaest","type":"user","name":"hsaest"},"name":"Jian Xie","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:08:52.947Z","hidden":false},{"_id":"6a152c7cb57a1823d5708b71","name":"Tianhe Lin","hidden":false},{"_id":"6a152c7cb57a1823d5708b72","name":"Zilu Wang","hidden":false},{"_id":"6a152c7cb57a1823d5708b73","user":{"_id":"65ace92f64c9b93eca5c2bce","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65ace92f64c9b93eca5c2bce/pG0JRXH-8zEy0IoaEnMNw.jpeg","isPro":false,"fullname":"Yuting Ning","user":"nnnyt","type":"user","name":"nnnyt"},"name":"Yuting Ning","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:08:48.531Z","hidden":false},{"_id":"6a152c7cb57a1823d5708b74","name":"Yuekun Yao","hidden":false},{"_id":"6a152c7cb57a1823d5708b75","name":"Tianci Xue","hidden":false},{"_id":"6a152c7cb57a1823d5708b76","name":"Zhehao Zhang","hidden":false},{"_id":"6a152c7cb57a1823d5708b77","user":{"_id":"671002fd13203512e7b8f9e3","avatarUrl":"/avatars/313d8ea313ed300750cfdaaca44fdb6e.svg","isPro":false,"fullname":"Zhongyang Li","user":"Lzy01241010","type":"user","name":"Lzy01241010"},"name":"Zhongyang Li","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:08:46.634Z","hidden":false},{"_id":"6a152c7cb57a1823d5708b78","user":{"_id":"63e0a50242591dda0b9dca5c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63e0a50242591dda0b9dca5c/c7cBPEBWQDFYimfGnO_SI.png","isPro":false,"fullname":"Kai Zhang","user":"drogozhang","type":"user","name":"drogozhang"},"name":"Kai Zhang","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:08:50.414Z","hidden":false},{"_id":"6a152c7cb57a1823d5708b79","name":"Yufan Wu","hidden":false},{"_id":"6a152c7cb57a1823d5708b7a","name":"Shijie Chen","hidden":false},{"_id":"6a152c7cb57a1823d5708b7b","user":{"_id":"6500870f1e14749e84f8f887","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6500870f1e14749e84f8f887/wfvx4BZvh2OyW-vpq5jEy.jpeg","isPro":false,"fullname":"Boyu Gou","user":"BoyuNLP","type":"user","name":"BoyuNLP"},"name":"Boyu Gou","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:08:54.765Z","hidden":false},{"_id":"6a152c7cb57a1823d5708b7c","name":"Mingzhe Han","hidden":false},{"_id":"6a152c7cb57a1823d5708b7d","name":"Yifei Wang","hidden":false},{"_id":"6a152c7cb57a1823d5708b7e","name":"Vint Lee","hidden":false},{"_id":"6a152c7cb57a1823d5708b7f","name":"Xinpeng Wei","hidden":false},{"_id":"6a152c7cb57a1823d5708b80","name":"Xiangjun Wang","hidden":false},{"_id":"6a152c7cb57a1823d5708b81","name":"Yu Su","hidden":false},{"_id":"6a152c7cb57a1823d5708b82","name":"Huan Sun","hidden":false}],"publishedAt":"2026-05-22T00:00:00.000Z","submittedOnDailyAt":"2026-05-26T00:00:00.000Z","title":"QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks","submittedOnDailyBy":{"_id":"62d65139667051e0a29bffe7","avatarUrl":"/avatars/0252aa2bcd4cf1c8e4b87e5f164b6da5.svg","isPro":false,"fullname":"Jian Xie","user":"hsaest","type":"user","name":"hsaest"},"summary":"Deep research agents extend the role of search engines from retrieving keyword-matched pages to synthesizing knowledge, fundamentally changing how humans interact with information. However, frontier systems remain proprietary, while existing open agents often generalize poorly across different task types, leaving unclear how to train a broadly capable deep research agent. We release QUEST, a family of open models (ranging from 2B to 35B) that serve as general-purpose deep research agents designed to handle a wide range of long-horizon search tasks, with strong capabilities in fact seeking, citation grounding, and report synthesis. To build QUEST, we propose an effective training recipe combining mid-training, supervised fine-tuning, and reinforcement learning. Central to this recipe is a curated data synthesis pipeline based on unified rubric trees, which applies to different task types and enables synthesizing training data with verifiable rewards without human annotation. In addition, QUEST incorporates a built-in context management mechanism that enables effective long-horizon reasoning and knowledge synthesis. Using only 8K synthesized tasks, QUEST approaches or even surpasses frontier closed-source agents across eight deep research benchmarks spanning diverse task types, and achieves the best overall performance among recent open-weight agents. We released everything: models, data, and training scripts.","upvotes":18,"discussionId":"6a152c7cb57a1823d5708b83","projectPage":"https://osu-nlp-group.github.io/QUEST/","githubRepo":"https://github.com/OSU-NLP-Group/QUEST","githubRepoAddedBy":"user","ai_summary":"QUEST is an open-family of deep research agents trained with synthesized data and reinforcement learning to perform well across diverse long-horizon search tasks.","ai_keywords":["supervised fine-tuning","reinforcement learning","data synthesis pipeline","rubric trees","context management mechanism","long-horizon reasoning","knowledge synthesis"],"githubStars":10,"organization":{"_id":"6127b4827dcb442c226129da","name":"osunlp","fullname":"OSU NLP Group","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6477a323dbc2a416f8b852b3/oiPPBo_knuDrz0YN9slKj.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"62d65139667051e0a29bffe7","avatarUrl":"/avatars/0252aa2bcd4cf1c8e4b87e5f164b6da5.svg","isPro":false,"fullname":"Jian Xie","user":"hsaest","type":"user"},{"_id":"643d4b27482011f5f2be350f","avatarUrl":"/avatars/0b05be4e0ffd0e7a60aa971a99546c6a.svg","isPro":false,"fullname":"xuetianci","user":"xuetianci99","type":"user"},{"_id":"671002fd13203512e7b8f9e3","avatarUrl":"/avatars/313d8ea313ed300750cfdaaca44fdb6e.svg","isPro":false,"fullname":"Zhongyang Li","user":"Lzy01241010","type":"user"},{"_id":"60a4ebfbaa9320dbbe69e37c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60a4ebfbaa9320dbbe69e37c/QLaEohXCWaUy8YX3wKQ_w.jpeg","isPro":false,"fullname":"Yiheng Shu","user":"yhshu","type":"user"},{"_id":"67169e3fd720d7d51e36a67e","avatarUrl":"/avatars/91113e5520009a4ed709bc62b96f2150.svg","isPro":false,"fullname":"WeijianQi","user":"WeijianQi1999","type":"user"},{"_id":"632ed3e534c562cb1d328fa7","avatarUrl":"/avatars/e29e47e734740787ea40c729eec146b5.svg","isPro":false,"fullname":"zeyi liao","user":"lzy337","type":"user"},{"_id":"643f9e2288d9d4488fd81c52","avatarUrl":"/avatars/e589c9cbd47022883cf33d7555bee89c.svg","isPro":false,"fullname":"Tinghui Zhu","user":"DarthZhu","type":"user"},{"_id":"64f08fc27a581687ff8deff9","avatarUrl":"/avatars/fa1bfc2f774a99d5a809776709ce8cb0.svg","isPro":false,"fullname":"Steven Shi","user":"Stevencivor","type":"user"},{"_id":"6500870f1e14749e84f8f887","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6500870f1e14749e84f8f887/wfvx4BZvh2OyW-vpq5jEy.jpeg","isPro":false,"fullname":"Boyu Gou","user":"BoyuNLP","type":"user"},{"_id":"63e0a50242591dda0b9dca5c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63e0a50242591dda0b9dca5c/c7cBPEBWQDFYimfGnO_SI.png","isPro":false,"fullname":"Kai Zhang","user":"drogozhang","type":"user"},{"_id":"66f0e2fbf41edbce00abbbbd","avatarUrl":"/avatars/9c3bf0e166d2983872da319b2e29a367.svg","isPro":false,"fullname":"Yikai Zhang","user":"ykzhang721","type":"user"},{"_id":"646dbd9fdf618b303b4073ef","avatarUrl":"/avatars/5b3bf8373c8ad403b7729e02e4fed224.svg","isPro":false,"fullname":"THLin","user":"nishinonana","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6127b4827dcb442c226129da","name":"osunlp","fullname":"OSU NLP Group","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6477a323dbc2a416f8b852b3/oiPPBo_knuDrz0YN9slKj.png"}}">

Cite arxiv.org/abs/2605.24218 in a model README.md to link it from this page.

Cite arxiv.org/abs/2605.24218 in a dataset README.md to link it from this page.