DexJoCo is a toolkit and benchmark for task-oriented dexterous manipulation based on MuJoCo with full-stack infrastructure, including open-sourced hardware design, data collection pipeline, human-collected trajectories, and long-horizon tool-use tasks.</p>\n","updatedAt":"2026-05-18T02:52:06.018Z","author":{"_id":"649ecf9827145c4463240177","avatarUrl":"/avatars/27696cf31790a3d58d8be2e0c983800e.svg","fullname":"Lue Fan","name":"Abyssaledge","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.878700852394104},"editors":["Abyssaledge"],"editorAvatarUrls":["/avatars/27696cf31790a3d58d8be2e0c983800e.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.16257","authors":[{"_id":"6a0a759875184a0d71e02610","user":{"_id":"6875f3e32944cd16067c2522","avatarUrl":"/avatars/db1231ac047fa5d72be60f04c29189a8.svg","isPro":false,"fullname":"whw","user":"gothicwhw","type":"user","name":"gothicwhw"},"name":"Hanwen Wang","status":"claimed_verified","statusLastChangedAt":"2026-05-18T09:40:23.112Z","hidden":false},{"_id":"6a0a759875184a0d71e02611","name":"Weizhi Zhao","hidden":false},{"_id":"6a0a759875184a0d71e02612","name":"Xiangyu Wang","hidden":false},{"_id":"6a0a759875184a0d71e02613","user":{"_id":"634e4120038b5879133552f5","avatarUrl":"/avatars/34ec861b4bbf1aecf927a7d6e726c7a4.svg","isPro":true,"fullname":"Siyuan","user":"SiyuanH","type":"user","name":"SiyuanH"},"name":"Siyuan Huang","status":"claimed_verified","statusLastChangedAt":"2026-05-18T07:46:51.266Z","hidden":false},{"_id":"6a0a759875184a0d71e02614","name":"He Lin","hidden":false},{"_id":"6a0a759875184a0d71e02615","name":"Boyuan Zheng","hidden":false},{"_id":"6a0a759875184a0d71e02616","name":"Rongtao Xu","hidden":false},{"_id":"6a0a759875184a0d71e02617","name":"Gang Wang","hidden":false},{"_id":"6a0a759875184a0d71e02618","name":"Yao Mu","hidden":false},{"_id":"6a0a759875184a0d71e02619","name":"He Wang","hidden":false},{"_id":"6a0a759875184a0d71e0261a","name":"Lue Fan","hidden":false},{"_id":"6a0a759875184a0d71e0261b","name":"Hongsheng Li","hidden":false},{"_id":"6a0a759875184a0d71e0261c","name":"Zhaoxiang Zhang","hidden":false},{"_id":"6a0a759875184a0d71e0261d","name":"Tieniu Tan","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/649ecf9827145c4463240177/C_YK4GcxZquUDIt2ODKS0.mp4"],"publishedAt":"2026-05-15T00:00:00.000Z","submittedOnDailyAt":"2026-05-18T00:00:00.000Z","title":"DexJoCo: A Benchmark and Toolkit for Task-Oriented Dexterous Manipulation on MuJoCo","submittedOnDailyBy":{"_id":"649ecf9827145c4463240177","avatarUrl":"/avatars/27696cf31790a3d58d8be2e0c983800e.svg","isPro":false,"fullname":"Lue Fan","user":"Abyssaledge","type":"user","name":"Abyssaledge"},"summary":"Achieving human-level manipulation requires dexterous robotic hands capable of complex object interactions. Advancing such capabilities further demands standardized benchmarks for systematic evaluation. However, existing dexterous benchmarks lack tasks that reflect the unique manipulation capabilities of dexterous hands over parallel grippers, as well as comprehensive evaluation pipelines. In this paper, we present DexJoCo, a benchmark and toolkit for task-oriented dexterous manipulation, comprising 11 functionally grounded tasks that evaluate tool-use, bimanual coordination, long-horizon execution, and reasoning. We develop a low-cost data collection system and collect 1.1K trajectories across these tasks, with support for domain randomization to assess robustness. We benchmark modern models under diverse settings, including visual and dynamics randomization, multi-task training, and action-head adaptation. Through extensive empirical analysis, we identify several important insights and common limitations of current policies in dexterous manipulation, highlighting key challenges for future research in dexterous hand robot learning. Project page available at: https://dexjoco.github.io","upvotes":46,"discussionId":"6a0a759875184a0d71e0261e","projectPage":"https://dexjoco.github.io/","githubRepo":"https://github.com/brave-eai/dexjoco","githubRepoAddedBy":"user","ai_summary":"DexJoCo presents a benchmark and toolkit for dexterous manipulation with 11 functional tasks evaluating tool-use, bimanual coordination, and long-horizon execution, along with a low-cost data collection system and comprehensive model evaluation.","ai_keywords":["dexterous manipulation","benchmark","toolkit","functionally grounded tasks","tool-use","bimanual coordination","long-horizon execution","domain randomization","visual randomization","dynamics randomization","multi-task training","action-head adaptation"],"githubStars":39},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"649ecf9827145c4463240177","avatarUrl":"/avatars/27696cf31790a3d58d8be2e0c983800e.svg","isPro":false,"fullname":"Lue Fan","user":"Abyssaledge","type":"user"},{"_id":"66a889cbe0420835a19e6f12","avatarUrl":"/avatars/a0f379be9d847ca4577cd9f39ca41089.svg","isPro":false,"fullname":"Weizhi Zhao","user":"WeizhiZhao","type":"user"},{"_id":"634e4120038b5879133552f5","avatarUrl":"/avatars/34ec861b4bbf1aecf927a7d6e726c7a4.svg","isPro":true,"fullname":"Siyuan","user":"SiyuanH","type":"user"},{"_id":"67cad8595aec2f093f2f46eb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3KHHj2woK64s3ocjyqvcu.png","isPro":false,"fullname":"Shayn Wang","user":"Soul-Reflecting","type":"user"},{"_id":"67c85fae58b004bdce6bd66a","avatarUrl":"/avatars/f007681b85de1191475a40115d0b58bf.svg","isPro":false,"fullname":"North.Forest","user":"North137","type":"user"},{"_id":"68412a6554d7c6b3f96aad07","avatarUrl":"/avatars/db4b3ad694b964279740f391cb61acd4.svg","isPro":false,"fullname":"Yubuzi","user":"Yubuzi20296","type":"user"},{"_id":"679897b49f6d200ac8120dc0","avatarUrl":"/avatars/e73626f89ca5b7d3c4168148c54cf8ba.svg","isPro":false,"fullname":"lisi","user":"liasdfe","type":"user"},{"_id":"673b10a75373479538b18bf4","avatarUrl":"/avatars/23dbaf07f9df644ec429b576ddac3d9a.svg","isPro":false,"fullname":"Fudong Ge","user":"fudongge","type":"user"},{"_id":"652f68596504631c10fe92a5","avatarUrl":"/avatars/91fac632a94de853bc7100b20a950cb4.svg","isPro":false,"fullname":"yaoyao","user":"yaoyao-jpg","type":"user"},{"_id":"669a1c1c4ac5ad025056c1b0","avatarUrl":"/avatars/a7d583179ce00073cfb967fcf764d75d.svg","isPro":false,"fullname":"Ruizhi Li","user":"KlayLi501","type":"user"},{"_id":"691d7b88208ffcf507951fd4","avatarUrl":"/avatars/f527d579062fbd2abcc0e8a4e1b885e1.svg","isPro":false,"fullname":"haoran xu","user":"xhrabc5678","type":"user"},{"_id":"6875f3e32944cd16067c2522","avatarUrl":"/avatars/db1231ac047fa5d72be60f04c29189a8.svg","isPro":false,"fullname":"whw","user":"gothicwhw","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.16257.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2605.16257 in a model README.md to link it from this page.

Cite arxiv.org/abs/2605.16257 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2605.16257 in a Space README.md to link it from this page.