Our framework synthesizes stable, functionally diverse grasps for unseen objects by decoupling functional intent from geometry. By learning a latent manifold in a canonical workspace, we achieve zero-shot generalization across multiple contact topologies.</p>\n","updatedAt":"2026-08-21T13:10:38.706Z","author":{"_id":"6687eb3ec056c55864f09152","avatarUrl":"/avatars/48efa05243507fbd1fe7a2f2139a19eb.svg","fullname":"Julien Mérand","name":"JulienMERAND","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8668121695518494},"editors":["JulienMERAND"],"editorAvatarUrls":["/avatars/48efa05243507fbd1fe7a2f2139a19eb.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.19776","authors":[{"_id":"6a883045404cbc83ad9f08b0","user":{"_id":"6687eb3ec056c55864f09152","avatarUrl":"/avatars/48efa05243507fbd1fe7a2f2139a19eb.svg","isPro":false,"fullname":"Julien Mérand","user":"JulienMERAND","type":"user","name":"JulienMERAND"},"name":"Julien Merand","status":"claimed_verified","statusLastChangedAt":"2026-08-21T12:56:54.782Z","hidden":false},{"_id":"6a883045404cbc83ad9f08b1","name":"Boris Meden","hidden":false},{"_id":"6a883045404cbc83ad9f08b2","name":"Liming Chen","hidden":false},{"_id":"6a883045404cbc83ad9f08b3","name":"Mathieu Grossard","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/6687eb3ec056c55864f09152/U9TvJ_VzB9JYbzzgzR-3e.png"],"publishedAt":"2026-08-20T00:00:00.000Z","submittedOnDailyAt":"2026-08-21T00:00:00.000Z","title":"CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learning","submittedOnDailyBy":{"_id":"6687eb3ec056c55864f09152","avatarUrl":"/avatars/48efa05243507fbd1fe7a2f2139a19eb.svg","isPro":false,"fullname":"Julien Mérand","user":"JulienMERAND","type":"user","name":"JulienMERAND"},"summary":"Current dexterous grasp planners primarily optimize for physical stability, focusing on whether an object can be grasped rather than how it should be grasped to support downstream functional tasks. However, conditioning grasp synthesis on specific human grasp taxonomies typically requires prohibitively expensive, object-annotated datasets. To address these limitations, we propose CoToGrasp, a novel generative framework that synthesizes diverse, stable grasps strictly conditioned on specific contact topologies. To bypass the data collection bottleneck, CoToGrasp is trained entirely in an object-agnostic manner. We introduce a feature-based canonical workspace that projects local object features into a unified gripper-centric domain, effectively decoupling the semantic functional intent from the arbitrary object geometry. By learning the intrinsic contact manifold of the gripper within this workspace, our model achieves zero-shot generalization to unseen objects at inference. Extensive evaluations on the large-scale DexGraspNet dataset demonstrate that CoToGrasp achieves state-of-the-art performance, outperforming existing taxonomy-guided planners. Finally, we demonstrate the physical viability and kinematic feasibility of our synthesized contact topologies on a physical robot platform. Code is available on our project website https://cea-list.github.io/cotograspweb/ .","upvotes":1,"discussionId":"6a883045404cbc83ad9f08b4","projectPage":"https://cea-list.github.io/cotograspweb/","githubRepo":"https://github.com/CEA-LIST/CoToGrasp","githubRepoAddedBy":"user","ai_summary":"CoToGrasp is a generative framework that synthesizes diverse, stable grasps conditioned on specific contact topologies using an object-agnostic, gripper-centric workspace for zero-shot generalization.","ai_keywords":["generative framework","contact topologies","object-agnostic","canonical workspace","contact manifold","zero-shot generalization","taxonomy-guided planners"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":3},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6687eb3ec056c55864f09152","avatarUrl":"/avatars/48efa05243507fbd1fe7a2f2139a19eb.svg","isPro":false,"fullname":"Julien Mérand","user":"JulienMERAND","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.19776.md","query":{}}">
CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learning
Abstract
CoToGrasp is a generative framework that synthesizes diverse, stable grasps conditioned on specific contact topologies using an object-agnostic, gripper-centric workspace for zero-shot generalization.
Current dexterous grasp planners primarily optimize for physical stability, focusing on whether an object can be grasped rather than how it should be grasped to support downstream functional tasks. However, conditioning grasp synthesis on specific human grasp taxonomies typically requires prohibitively expensive, object-annotated datasets. To address these limitations, we propose CoToGrasp, a novel generative framework that synthesizes diverse, stable grasps strictly conditioned on specific contact topologies. To bypass the data collection bottleneck, CoToGrasp is trained entirely in an object-agnostic manner. We introduce a feature-based canonical workspace that projects local object features into a unified gripper-centric domain, effectively decoupling the semantic functional intent from the arbitrary object geometry. By learning the intrinsic contact manifold of the gripper within this workspace, our model achieves zero-shot generalization to unseen objects at inference. Extensive evaluations on the large-scale DexGraspNet dataset demonstrate that CoToGrasp achieves state-of-the-art performance, outperforming existing taxonomy-guided planners. Finally, we demonstrate the physical viability and kinematic feasibility of our synthesized contact topologies on a physical robot platform. Code is available on our project website https://cea-list.github.io/cotograspweb/ .
Community
Our framework synthesizes stable, functionally diverse grasps for unseen objects by decoupling functional intent from geometry. By learning a latent manifold in a canonical workspace, we achieve zero-shot generalization across multiple contact topologies.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.19776 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.19776 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.19776 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.