LingBot-VLA 2.0<br>From Foundation to Application: Improving VLA Models in Practice</p>\n","updatedAt":"2026-07-08T07:46:49.624Z","author":{"_id":"69773d4ee6878183fb90a8c7","avatarUrl":"/avatars/3e9e4081e3beaf3f69c380387b8ee4c2.svg","fullname":"Wei Wu","name":"Weiww99","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":2,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6796033382415771},"editors":["Weiww99"],"editorAvatarUrls":["/avatars/3e9e4081e3beaf3f69c380387b8ee4c2.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.06403","authors":[{"_id":"6a4dff5d25849b193a834c9b","name":"Wei Wu","hidden":false},{"_id":"6a4dff5d25849b193a834c9c","name":"Fangjing Wang","hidden":false},{"_id":"6a4dff5d25849b193a834c9d","name":"Fan Lu","hidden":false},{"_id":"6a4dff5d25849b193a834c9e","name":"He Sun","hidden":false},{"_id":"6a4dff5d25849b193a834c9f","name":"Shi Liu","hidden":false},{"_id":"6a4dff5d25849b193a834ca0","name":"Yunnan Wang","hidden":false},{"_id":"6a4dff5d25849b193a834ca1","name":"Yibin Yan","hidden":false},{"_id":"6a4dff5d25849b193a834ca2","name":"Yong Wang","hidden":false},{"_id":"6a4dff5d25849b193a834ca3","name":"Shuailei Ma","hidden":false},{"_id":"6a4dff5d25849b193a834ca4","name":"Xinyang Wang","hidden":false},{"_id":"6a4dff5d25849b193a834ca5","name":"Yibin Liu","hidden":false},{"_id":"6a4dff5d25849b193a834ca6","name":"Shuai Yang","hidden":false},{"_id":"6a4dff5d25849b193a834ca7","name":"Tianxiang Zhou","hidden":false},{"_id":"6a4dff5d25849b193a834ca8","name":"Kejia Zhang","hidden":false},{"_id":"6a4dff5d25849b193a834ca9","name":"Lei Zhou","hidden":false},{"_id":"6a4dff5d25849b193a834caa","name":"Cheng Su","hidden":false},{"_id":"6a4dff5d25849b193a834cab","name":"Nan Xue","hidden":false},{"_id":"6a4dff5d25849b193a834cac","name":"Bin Tan","hidden":false},{"_id":"6a4dff5d25849b193a834cad","name":"Han Zhang","hidden":false},{"_id":"6a4dff5d25849b193a834cae","name":"Youchao Zhang","hidden":false},{"_id":"6a4dff5d25849b193a834caf","name":"Fei Liao","hidden":false},{"_id":"6a4dff5d25849b193a834cb0","name":"Xing Zhu","hidden":false},{"_id":"6a4dff5d25849b193a834cb1","name":"Yujun Shen","hidden":false},{"_id":"6a4dff5d25849b193a834cb2","name":"Kecheng Zheng","hidden":false}],"publishedAt":"2026-07-07T00:00:00.000Z","submittedOnDailyAt":"2026-07-08T00:00:00.000Z","title":"From Foundation to Application: Improving VLA Models in Practice","submittedOnDailyBy":{"_id":"69773d4ee6878183fb90a8c7","avatarUrl":"/avatars/3e9e4081e3beaf3f69c380387b8ee4c2.svg","isPro":false,"fullname":"Wei Wu","user":"Weiww99","type":"user","name":"Weiww99"},"summary":"Despite recent progress of VLA foundation models, the disparity between laboratory conditions and real-world applications continues to impede their practical implementation. To bridge this gap, we present LingBot-VLA 2.0, which advances LingBot-VLA through improvements in three functional domains. (1) Generalization across tasks and embodiments. Compared to the previous version, we revamp the data processing pipeline and curate around 60,000 hours of data for pretraining, including 50,000 hours of robot trajectories spanning 20 robot configurations and 10,000 hours of egocentric human videos. (2) Expanded action space in addition to dual-arm hardware platforms. In particular, our system accommodates degrees of freedom for the heads, waists, mobile bases, and dexterous hands, thereby empowering the robots to tackle more complex tasks in practical scenarios. (3) Predictive dynamics modeling for improved temporal reasoning. Specifically, we formulate future prediction as a proxy task, facilitated by a video representation model for semantic priors and a depth estimation model for geometric cues. Evaluations on the GM-100 benchmark, conducted in a generalist setting, validate the beneficial impact of these proposed modifications. Furthermore, benefiting from the expanded pretraining data that covers whole-body degrees of freedom, LingBot-VLA-2.0 demonstrates strong cross-embodiment long-horizon mobile manipulation capability across the two robotic platforms.","upvotes":10,"discussionId":"6a4dff5e25849b193a834cb3","projectPage":"https://technology.robbyant.com/lingbot-vla-v2","githubRepo":"https://github.com/robbyant/lingbot-vla-v2","githubRepoAddedBy":"user","ai_summary":"LingBot-VLA 2.0 enhances generalization across tasks and embodiments through expanded data preprocessing and training on diverse robot configurations, extends action space to include whole-body degrees of freedom for complex manipulation tasks, and incorporates predictive dynamics modeling using video representation and depth estimation for improved temporal reasoning.","ai_keywords":["VLA foundation models","data processing pipeline","robot trajectories","egocentric human videos","action space","degrees of freedom","video representation model","depth estimation model","predictive dynamics modeling","temporal reasoning","GM-100 benchmark","cross-embodiment long-horizon mobile manipulation"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":197,"organization":{"_id":"69709f892cd08371c1011a2e","name":"robbyant","fullname":"Robbyant","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67aeffda7330db26f93cd62f/ZTuImney4XzRmBHyUL47F.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"67aeffda7330db26f93cd62f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/67aeffda7330db26f93cd62f/ktdFt5_F05qNJ-dnD_Fk3.jpeg","isPro":false,"fullname":"jiangbonadia","user":"NadiaJiang","type":"user"},{"_id":"63c1699e40a26dd2db32400d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63c1699e40a26dd2db32400d/3N0-Zp8igv8-52mXAdiiq.jpeg","isPro":false,"fullname":"Chroma","user":"Chroma111","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":"687363d49a81c7dcbcfa2d84","avatarUrl":"/avatars/5d943a5c811ed931c3fdcfee19253049.svg","isPro":false,"fullname":"jj","user":"realman123","type":"user"},{"_id":"6186ddf6a7717cb375090c01","avatarUrl":"/avatars/716b6a7d1094c8036b2a8a7b9063e8aa.svg","isPro":true,"fullname":"Julien BLANCHON","user":"blanchon","type":"user"},{"_id":"6721acf7bbf9703bc285e840","avatarUrl":"/avatars/487f5cae476758d318465aa9ed103568.svg","isPro":false,"fullname":"Yujie Zhao","user":"HomieZ","type":"user"},{"_id":"61ca43f194afcb20ceb764d7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1640645587885-noauth.jpeg","isPro":false,"fullname":"Raul Garrido","user":"happybydefault","type":"user"},{"_id":"6881b53e8579760ec4199c3a","avatarUrl":"/avatars/94041eea0fe43b4fc056d6380e8f6633.svg","isPro":false,"fullname":"StreamFormer","user":"StreamFormer","type":"user"},{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"},{"_id":"6395b7642adf062a0b252d4e","avatarUrl":"/avatars/aa9438c4ed337836a49e22ffd7187359.svg","isPro":false,"fullname":"Wangfangjing","user":"Yioutpi","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"69709f892cd08371c1011a2e","name":"robbyant","fullname":"Robbyant","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67aeffda7330db26f93cd62f/ZTuImney4XzRmBHyUL47F.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.06403.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.06403 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.06403 in a Space README.md to link it from this page.