Website: <a href=\"https://robodojo-benchmark.com/\" rel=\"nofollow\">https://robodojo-benchmark.com/</a><br>arXiv: <a href=\"https://arxiv.org/abs/2607.04434\" rel=\"nofollow\">https://arxiv.org/abs/2607.04434</a><br>Leaderboard: <a href=\"https://robodojo-benchmark.com/LeaderBoard\" rel=\"nofollow\">https://robodojo-benchmark.com/LeaderBoard</a><br>Benchmark code: <a href=\"https://github.com/RoboDojo-Benchmark/RoboDojo\" rel=\"nofollow\">https://github.com/RoboDojo-Benchmark/RoboDojo</a><br>XPolicyLab code: <a href=\"https://github.com/XPolicyLab/XPolicyLab\" rel=\"nofollow\">https://github.com/XPolicyLab/XPolicyLab</a><br>Community: <a href=\"https://robodojo-benchmark.com/community\" rel=\"nofollow\">https://robodojo-benchmark.com/community</a></p>\n","updatedAt":"2026-07-09T02:48:48.892Z","author":{"_id":"65b37a9b06d8b55123ef8921","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b37a9b06d8b55123ef8921/CT5tLwezjXct1eTszA8sO.jpeg","fullname":"Tianxing Chen","name":"TianxingChen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":17,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.5292746424674988},"editors":["TianxingChen"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/65b37a9b06d8b55123ef8921/CT5tLwezjXct1eTszA8sO.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.04434","authors":[{"_id":"6a4dcd4e25849b193a834c06","name":"Tianxing Chen","hidden":false},{"_id":"6a4dcd4e25849b193a834c07","name":"Yue Chen","hidden":false},{"_id":"6a4dcd4e25849b193a834c08","name":"Zixuan Li","hidden":false},{"_id":"6a4dcd4e25849b193a834c09","name":"Junyuan Tang","hidden":false},{"_id":"6a4dcd4e25849b193a834c0a","name":"Kailun Su","hidden":false},{"_id":"6a4dcd4e25849b193a834c0b","name":"Haoran Lu","hidden":false},{"_id":"6a4dcd4e25849b193a834c0c","name":"Weijie Wan","hidden":false},{"_id":"6a4dcd4e25849b193a834c0d","name":"Baijun Chen","hidden":false},{"_id":"6a4dcd4e25849b193a834c0e","name":"Songling Liu","hidden":false},{"_id":"6a4dcd4e25849b193a834c0f","name":"Haowen Yan","hidden":false},{"_id":"6a4dcd4e25849b193a834c10","name":"Honghao Su","hidden":false},{"_id":"6a4dcd4e25849b193a834c11","name":"Zhiyang Dou","hidden":false},{"_id":"6a4dcd4e25849b193a834c12","name":"Kaixuan Wang","hidden":false},{"_id":"6a4dcd4e25849b193a834c13","name":"Dandan Zhang","hidden":false},{"_id":"6a4dcd4e25849b193a834c14","name":"Yunze Liu","hidden":false},{"_id":"6a4dcd4e25849b193a834c15","name":"Yan Qin","hidden":false},{"_id":"6a4dcd4e25849b193a834c16","name":"Qiwei Liang","hidden":false},{"_id":"6a4dcd4e25849b193a834c17","name":"Qiwei Wu","hidden":false},{"_id":"6a4dcd4e25849b193a834c18","name":"Zijian Lin","hidden":false},{"_id":"6a4dcd4e25849b193a834c19","name":"Wenwei Lin","hidden":false},{"_id":"6a4dcd4e25849b193a834c1a","name":"Yuran Wang","hidden":false},{"_id":"6a4dcd4e25849b193a834c1b","name":"Minghua He","hidden":false},{"_id":"6a4dcd4e25849b193a834c1c","name":"Tianshu Wu","hidden":false},{"_id":"6a4dcd4e25849b193a834c1d","name":"Ruihai Wu","hidden":false},{"_id":"6a4dcd4e25849b193a834c1e","name":"Jingquan Zhou","hidden":false},{"_id":"6a4dcd4e25849b193a834c1f","name":"Kai-Chong Lei","hidden":false},{"_id":"6a4dcd4e25849b193a834c20","name":"Haibao Yu","hidden":false},{"_id":"6a4dcd4e25849b193a834c21","name":"Yuanfeng Ji","hidden":false},{"_id":"6a4dcd4e25849b193a834c22","name":"Weiyang Jin","hidden":false},{"_id":"6a4dcd4e25849b193a834c23","name":"Guanyu Lin","hidden":false},{"_id":"6a4dcd4e25849b193a834c24","name":"Xiaofan Li","hidden":false},{"_id":"6a4dcd4e25849b193a834c25","name":"Qi Xiong","hidden":false},{"_id":"6a4dcd4e25849b193a834c26","name":"Renjing Xu","hidden":false},{"_id":"6a4dcd4e25849b193a834c27","name":"Zhongyu Li","hidden":false},{"_id":"6a4dcd4e25849b193a834c28","name":"Wenhao Chai","hidden":false},{"_id":"6a4dcd4e25849b193a834c29","name":"Enze Xie","hidden":false},{"_id":"6a4dcd4e25849b193a834c2a","name":"Ziwei Wang","hidden":false},{"_id":"6a4dcd4e25849b193a834c2b","name":"Yao Mu","hidden":false},{"_id":"6a4dcd4e25849b193a834c2c","name":"Hao Dong","hidden":false},{"_id":"6a4dcd4e25849b193a834c2d","name":"Wojciech Matusik","hidden":false},{"_id":"6a4dcd4e25849b193a834c2e","name":"Mingyu Ding","hidden":false},{"_id":"6a4dcd4e25849b193a834c2f","name":"Wenbo Ding","hidden":false},{"_id":"6a4dcd4e25849b193a834c30","name":"Ping Luo","hidden":false},{"_id":"6a4dcd4e25849b193a834c31","name":"Masayoshi Tomizuka","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/65b37a9b06d8b55123ef8921/M-5QhZgUyukN0I4dEpyL8.mp4"],"publishedAt":"2026-07-07T00:00:00.000Z","submittedOnDailyAt":"2026-07-09T00:00:00.000Z","title":"RoboDojo: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies","submittedOnDailyBy":{"_id":"65b37a9b06d8b55123ef8921","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b37a9b06d8b55123ef8921/CT5tLwezjXct1eTszA8sO.jpeg","isPro":false,"fullname":"Tianxing Chen","user":"TianxingChen","type":"user","name":"TianxingChen"},"summary":"Generalist robot manipulation policies have advanced rapidly, yet existing benchmarks remain limited in systematically evaluating their capabilities. Many rely on simple, short-horizon, or skill-narrow tasks with limited capability coverage, and are often conducted only in simulation or only in the real world. Simulation enables scalable feedback but misses physical deployment challenges, while real-world evaluation is costly, time-consuming, and difficult to reproduce. We introduce RoboDojo, a unified sim-and-real benchmark for comprehensive evaluation of generalist robot manipulation policies. RoboDojo includes 42 simulation tasks and 18 real-world tasks covering diverse and complementary manipulation capabilities. The simulation benchmark evaluates five dimensions: generalization, memory, precision, long-horizon execution, and open-vocabulary instruction following, while the real-world benchmark exposes policies to challenging physical-world deployment conditions. RoboDojo supports scalable evaluation through heterogeneous parallel simulation in Isaac Sim and provides RoboDojo-RealEval, a reproducible real-world evaluation system with remote cloud access, standardized hardware, scene reset, evaluation protocol, and deployment interface. Together with XPolicyLab, policies can be integrated once and evaluated across simulation and real-world settings with minimal adaptation. We integrate 30 policies into XPolicyLab and evaluate them on RoboDojo, establishing a public leaderboard and systematic analysis of current policy performance. The website is available at http://robodojo-benchmark.com/.","upvotes":4,"discussionId":"6a4dcd4f25849b193a834c32","projectPage":"https://robodojo-benchmark.com/","githubRepo":"https://github.com/RoboDojo-Benchmark/RoboDojo","githubRepoAddedBy":"user","ai_summary":"RoboDojo presents a unified sim-and-real benchmark for evaluating generalist robot manipulation policies across diverse tasks and evaluation dimensions.","ai_keywords":[""],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":92,"organization":{"_id":"6a4a92e27391729bb22ddf43","name":"RoboDojo-Benchmark","fullname":"RoboDojo-Benchmark","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67cc5e0232aeea9209d35033/evaqVzuHh_3XFQ5HRSxo5.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"65b37a9b06d8b55123ef8921","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b37a9b06d8b55123ef8921/CT5tLwezjXct1eTszA8sO.jpeg","isPro":false,"fullname":"Tianxing Chen","user":"TianxingChen","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":"642e97dbc1b0f8e4e76c2b30","avatarUrl":"/avatars/60adf4470baf12d5687d53a6c3299bcd.svg","isPro":false,"fullname":"james curry","user":"ainbo","type":"user"},{"_id":"645d7f107c7258d904e82749","avatarUrl":"/avatars/a4e9d47b281f18616c522c1a8b8ee7e5.svg","isPro":false,"fullname":"HuichiZhou","user":"Zhouhc","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6a4a92e27391729bb22ddf43","name":"RoboDojo-Benchmark","fullname":"RoboDojo-Benchmark","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67cc5e0232aeea9209d35033/evaqVzuHh_3XFQ5HRSxo5.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.04434.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.04434 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.04434 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.