📄 arXiv: <a href=\"https://lnkd.in/gHbJeZ-E\" rel=\"nofollow\">https://lnkd.in/gHbJeZ-E</a><br>🌐 Project page: <a href=\"https://lnkd.in/g3dJY-BU\" rel=\"nofollow\">https://lnkd.in/g3dJY-BU</a></p>\n<p>TL;DR: Large-scale robot datasets span diverse camera viewpoints, making policy learning harder. Give a VLA robot-centric pointmaps as input, and it consistently performs better with just one extra encoder and one element-wise addition.</p>\n","updatedAt":"2026-07-20T14:01:38.225Z","author":{"_id":"67061bf00cc840f4462a0705","avatarUrl":"/avatars/0fab53fa981f166483aa8ad61172d32d.svg","fullname":"Dongyoon","name":"godnpeter","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8486666679382324},"editors":["godnpeter"],"editorAvatarUrls":["/avatars/0fab53fa981f166483aa8ad61172d32d.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.11498","authors":[{"_id":"6a59bbce6c2e371e6ca38248","name":"Byungkun Lee","hidden":false},{"_id":"6a59bbce6c2e371e6ca38249","name":"Dongyoon Hwang","hidden":false},{"_id":"6a59bbce6c2e371e6ca3824a","name":"Dongjin Kim","hidden":false},{"_id":"6a59bbce6c2e371e6ca3824b","name":"Hojoon Lee","hidden":false},{"_id":"6a59bbce6c2e371e6ca3824c","name":"Minho Park","hidden":false},{"_id":"6a59bbce6c2e371e6ca3824d","name":"Jaegul Choo","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/67061bf00cc840f4462a0705/M2E2iEsoUju1Z_tSbIxkI.mp4"],"publishedAt":"2026-07-13T00:00:00.000Z","submittedOnDailyAt":"2026-07-20T00:00:00.000Z","title":"See like a Robot: Robot-Centric Pointmaps for Vision-Language-Action Models","submittedOnDailyBy":{"_id":"67061bf00cc840f4462a0705","avatarUrl":"/avatars/0fab53fa981f166483aa8ad61172d32d.svg","isPro":false,"fullname":"Dongyoon","user":"godnpeter","type":"user","name":"godnpeter"},"summary":"Vision-language-action (VLA) models predict robot actions from visual observations and language instructions. These actions are defined in the robot's own 3D coordinate frame, yet most VLAs observe the scene in the camera frame, creating a frame mismatch between where the scene is observed and where actions are defined. The mismatch is benign under a fixed viewpoint, where the policy can memorize a single observation-to-action mapping, but grows harder as large-scale datasets aggregate demonstrations across diverse camera setups and the policy must generalize this mapping across viewpoints. We address this mismatch with robot-centric pointmaps, images whose pixels store the 3D coordinates of scene points in the robot frame. Pointmaps provide robot-frame 3D geometry while preserving the dense H x W grid expected by pretrained 2D VLAs, so they integrate into existing VLAs with minimal architectural change. On RoboCasa, pointmaps improve both pi0.5 and SmolVLA and outperform representative camera-viewpoint and 3D-aware baselines. In real-robot experiments, their advantage over an RGB-only policy widens when the camera is moved to a placement unseen during training.","upvotes":3,"discussionId":"6a59bbcf6c2e371e6ca3824e","projectPage":"https://davian-robotics.github.io/pointmap/","organization":{"_id":"6475760c33192631bad2bb38","name":"kaist-ai","fullname":"KAIST AI","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6469949654873f0043b09c22/aaZFiyXe1qR-Dmy_xq67m.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"67061bf00cc840f4462a0705","avatarUrl":"/avatars/0fab53fa981f166483aa8ad61172d32d.svg","isPro":false,"fullname":"Dongyoon","user":"godnpeter","type":"user"},{"_id":"64587be872b60ae7a3817858","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64587be872b60ae7a3817858/BbdOOxOCEzWTvEpkWp8MM.png","isPro":false,"fullname":"Minbyul Jeong","user":"Minbyul","type":"user"},{"_id":"66be7f56780d735f17e94962","avatarUrl":"/avatars/8a7a13ca4f028bfa8c82df391ade22a6.svg","isPro":false,"fullname":"Gray Stanton","user":"GrStant","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6475760c33192631bad2bb38","name":"kaist-ai","fullname":"KAIST AI","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6469949654873f0043b09c22/aaZFiyXe1qR-Dmy_xq67m.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.11498.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.11498 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.11498 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.11498 in a Space README.md to link it from this page.