Humans glance at any object and instantly know where to act and how -- AFUN is an affordance foundation model that gives robots the same ability. From a single RGB-D image and a language command, it jointly predicts a task-conditional functional mask (where to interact) and a 3D post-contact motion curve (how to move), trained on one of the largest real-world affordance datasets to date, spanning robot, human, simulation, and 3D-scan sources. It takes a step toward open-world generalization, with state-of-the-art results across segmentation, contact-point, and 3D-motion benchmarks. And it deploys directly to a real robot, executing manipulation tasks like opening and grasping without any robot-specific fine-tuning.</p>\n","updatedAt":"2026-06-02T17:26:46.901Z","author":{"_id":"685b8cc0aa442e9a67b5f2a2","avatarUrl":"/avatars/ae0a745d8b8275e3b5ebfcf985c6aede.svg","fullname":"Zhaoning Wang","name":"Zhaoningw","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.910729169845581},"editors":["Zhaoningw"],"editorAvatarUrls":["/avatars/ae0a745d8b8275e3b5ebfcf985c6aede.svg"],"reactions":[],"isReport":false}},{"id":"6a1f8a85b47e980051378292","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":360,"isUserFollowing":false},"createdAt":"2026-06-03T01:59:33.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [BridgeACT: Bridging Human Demonstrations to Robot Actions via Unified Tool-Target Affordances](https://huggingface.co/papers/2604.23249) (2026)\n* [Afford-VLA: Action-Aligned Visual Planning via Internalized Affordance](https://huggingface.co/papers/2605.24203) (2026)\n* [Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation](https://huggingface.co/papers/2604.24681) (2026)\n* [Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments](https://huggingface.co/papers/2605.30280) (2026)\n* [Spatially Prompted Visual Trajectory Prediction for Egocentric Manipulation](https://huggingface.co/papers/2605.20085) (2026)\n* [Grasp as You Dream: Imitating Functional Grasping from Generated Human Demonstrations](https://huggingface.co/papers/2604.07517) (2026)\n* [AffordSim: A Scalable Data Generator and Benchmark for Affordance-Aware Robotic Manipulation](https://huggingface.co/papers/2604.11674) (2026)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"<p>This is an automated message from the <a href=\"https://huggingface.co/librarian-bots\">Librarian Bot</a>. I found the following papers similar to this paper. </p>\n<p>The following papers were recommended by the Semantic Scholar API </p>\n<ul>\n<li><a href=\"https://huggingface.co/papers/2604.23249\">BridgeACT: Bridging Human Demonstrations to Robot Actions via Unified Tool-Target Affordances</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2605.24203\">Afford-VLA: Action-Aligned Visual Planning via Internalized Affordance</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.24681\">Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2605.30280\">Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2605.20085\">Spatially Prompted Visual Trajectory Prediction for Egocentric Manipulation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.07517\">Grasp as You Dream: Imitating Functional Grasping from Generated Human Demonstrations</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2604.11674\">AffordSim: A Scalable Data Generator and Benchmark for Affordance-Aware Robotic Manipulation</a> (2026)</li>\n</ul>\n<p> Please give a thumbs up to this comment if you found it helpful!</p>\n<p> If you want recommendations for any Paper on Hugging Face checkout <a href=\"https://huggingface.co/spaces/librarian-bots/recommend_similar_papers\">this</a> Space</p>\n<p> You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: <code><span class=\"SVELTE_PARTIAL_HYDRATER contents\" data-target=\"UserMention\" data-props=\"{&quot;user&quot;:&quot;librarian-bot&quot;}\"><span class=\"inline-block\"><span class=\"contents\"><a href=\"/librarian-bot\">@<span class=\"underline\">librarian-bot</span></a></span> </span></span> recommend</code></p>\n","updatedAt":"2026-06-03T01:59:33.527Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":360,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7227720022201538},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.02551","authors":[{"_id":"6a1f02d0e292c1c78ecb1148","name":"Zhaoning Wang","hidden":false},{"_id":"6a1f02d0e292c1c78ecb1149","name":"Yi Zhong","hidden":false},{"_id":"6a1f02d0e292c1c78ecb114a","name":"Jiawei Fu","hidden":false},{"_id":"6a1f02d0e292c1c78ecb114b","name":"Henrik I. Christensen","hidden":false},{"_id":"6a1f02d0e292c1c78ecb114c","name":"Jun Gao","hidden":false}],"publishedAt":"2026-06-01T00:00:00.000Z","submittedOnDailyAt":"2026-06-02T00:00:00.000Z","title":"AFUN: Towards an Affordance Foundation Model for Functionality Understanding","submittedOnDailyBy":{"_id":"685b8cc0aa442e9a67b5f2a2","avatarUrl":"/avatars/ae0a745d8b8275e3b5ebfcf985c6aede.svg","isPro":false,"fullname":"Zhaoning Wang","user":"Zhaoningw","type":"user","name":"Zhaoningw"},"summary":"Affordance understanding bridges visual perception and physical action, serving as an explainable interface for robot manipulation in open and unstructured real-world environments. Yet, building an affordance foundation model that not only understands where and how the interaction should happen, but also generalizes across diverse environments, objects, and tasks, remains a long-standing research challenge. Existing methods typically address only part of this challenge, either localizing task-relevant regions without specifying executable motion, or predicting motion but with limited scalability. In this paper, we present ourmodel, a step towards an affordance foundation model for functionality understanding. From a single RGB-D observation and a language task description, ourmodel predicts a task-conditional functional mask (where to interact) and a 3D post-contact motion curve (how to interact). To support open-world generalization, we build a large-scale standardized data pipeline that converts heterogeneous robot, human, simulation, and real-world scan data into a shared affordance schema with language, masks, and object-centric 3D motion labels. We evaluate ourmodel from three aspects: for affordance segmentation, ourmodel outperforms all baselines by a large margin across 8 test sets from 4 benchmarks, improving mean gIoU/cIoU by +23.9/+26.3; for contact-point prediction, it predicts substantially more accurate points, with a 12.7--61.3% hit-rate gain over the best baseline; and for 3D motion, it achieves the best performance on all three test sets. ourmodel can be deployed for real-world robot manipulation without finetuning for robot embodiment or using task-specific heuristics, demonstrating the ability to adapt to open-world affordance tasks. Project page: https://www.zhaoningwang.com/AFUN","upvotes":6,"discussionId":"6a1f02d1e292c1c78ecb114d","projectPage":"https://www.zhaoningwang.com/AFUN/","githubRepo":"https://github.com/EricWang12/AFUN","githubRepoAddedBy":"user","ai_summary":"Affordance understanding model predicts functional masks and 3D motion curves from RGB-D observations and language descriptions, enabling generalizable robot manipulation across diverse environments.","ai_keywords":["affordance foundation model","task-conditional functional mask","3D post-contact motion curve","RGB-D observation","language task description","open-world generalization","functional mask","contact-point prediction","3D motion prediction"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":0,"organization":{"_id":"63df4874e742e86dc925d67c","name":"umich","fullname":"University of Michigan","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675577443573-63df328115266dd945fc01f4.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"685b8cc0aa442e9a67b5f2a2","avatarUrl":"/avatars/ae0a745d8b8275e3b5ebfcf985c6aede.svg","isPro":false,"fullname":"Zhaoning Wang","user":"Zhaoningw","type":"user"},{"_id":"646eebbacb6ea6e6b6e1f5f5","avatarUrl":"/avatars/a4adaf7a63c74ffd9115d0f78df38071.svg","isPro":false,"fullname":"Jiu Feng","user":"JanusVon","type":"user"},{"_id":"65e1c1cb7fd901af96753f93","avatarUrl":"/avatars/5853f9b93aca39074cec6efd29695d7d.svg","isPro":false,"fullname":"chen siyi","user":"csusupergear","type":"user"},{"_id":"654016a6cdc9c22e35a17829","avatarUrl":"/avatars/c81f84567656cce6c21b994b91de4100.svg","isPro":false,"fullname":"Kevin Xu","user":"KevinXu02","type":"user"},{"_id":"633b7a4b0d68f86e2d98de05","avatarUrl":"/avatars/5d48c171ddbcc7ca39bdc0d11c6224e4.svg","isPro":false,"fullname":"Jun Gao","user":"JungaoCanada","type":"user"},{"_id":"691ba6d93560b8c8cb6c4d45","avatarUrl":"/avatars/f57108af9258171fa01e18d0ba854905.svg","isPro":false,"fullname":"Yi Zhong","user":"yeszhong","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"63df4874e742e86dc925d67c","name":"umich","fullname":"University of Michigan","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675577443573-63df328115266dd945fc01f4.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.02551.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.02551 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.02551 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.02551 in a Space README.md to link it from this page.

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