Can every robot predict the same future collective state using only local observations and bandwidth-limited neighbor messages? We introduce CS-JEPA, a fully decentralized predictive architecture that transfers across unseen swarm topologies and sizes up to 108 robots while requiring fewer globally labeled episodes than future reconstruction. Further experiments demonstrate more accurate action-value estimation and more consistent closed-loop decisions across robots. See Version 2 of this paper, which includes additional experiments, on arXiv: <a href=\"https://arxiv.org/abs/2607.28443\" rel=\"nofollow\">https://arxiv.org/abs/2607.28443</a></p>\n","updatedAt":"2026-08-03T08:15:03.947Z","author":{"_id":"63654b82ad2c9e8b672d8a4b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667661623748-63654b82ad2c9e8b672d8a4b.jpeg","fullname":"Alan Gazzaev","name":"AlanRobotics","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":7,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8380286693572998},"editors":["AlanRobotics"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1667661623748-63654b82ad2c9e8b672d8a4b.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.28443","authors":[{"_id":"6a6c533f202e2d9e3ffdb893","user":{"_id":"63654b82ad2c9e8b672d8a4b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667661623748-63654b82ad2c9e8b672d8a4b.jpeg","isPro":false,"fullname":"Alan Gazzaev","user":"AlanRobotics","type":"user","name":"AlanRobotics"},"name":"Alan-Barsag Gazzaev","status":"claimed_verified","statusLastChangedAt":"2026-07-31T08:45:05.663Z","hidden":false},{"_id":"6a6c533f202e2d9e3ffdb894","name":"Alexey Garvilov","hidden":false},{"_id":"6a6c533f202e2d9e3ffdb895","name":"Sergey Muravyov","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/63654b82ad2c9e8b672d8a4b/MUFXKTZllPQPgO4tTaEqf.png"],"publishedAt":"2026-07-30T00:00:00.000Z","submittedOnDailyAt":"2026-08-03T00:00:00.000Z","title":"One Future, Every Robot: Label-Efficient Collective-State Prediction with Decentralized JEPA","submittedOnDailyBy":{"_id":"63654b82ad2c9e8b672d8a4b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667661623748-63654b82ad2c9e8b672d8a4b.jpeg","isPro":false,"fullname":"Alan Gazzaev","user":"AlanRobotics","type":"user","name":"AlanRobotics"},"summary":"Can every robot in a swarm predict the same future collective state from only local observations and bandwidth-limited messages? We formulate this as decentralized shared-state prediction and introduce Collective-State JEPA (CS-JEPA), a recurrent joint-embedding predictive architecture whose output at every robot represents one common future token field. At deployment, each robot uses a 16-frame local history and one 64-float recurrent message per directed edge; there is no global pooling, target encoder, episode clock, or recorded future action. After pretraining without downstream collective labels, frozen representations are evaluated with ridge probes fitted on 6, 12, or 24 globally labeled episodes. Against raw-future reconstruction with the same receiver anchor and deployment capacity but 9,607 additional training-only parameters, a prospectively registered five-seed follow-up improves prediction-error and inter-robot-agreement label-budget AUC on in-distribution, ring, mutual-kNN, and unseen-size families up to 108 robots. Every effect favors CS-JEPA in 5/5 outer seeds. In a separate sealed eight-seed follow-up, matched action-conditioned predictors receive each candidate four-step plan before producing receiver-local predictive representations. CS-JEPA reduces branch-value MSE by 45.5% and improves within-context candidate-score Pearson correlation by 0.1291, with both effects favorable in 8/8 seeds, including at unseen N=32. These results support common-future JEPA targets as a label-efficient primitive for decentralized swarm prediction under topology and size shift, with additional evidence of planning-relevant value estimation.","upvotes":3,"discussionId":"6a6c533f202e2d9e3ffdb896","organization":{"_id":"61b8da8ed48e2b6a4d2f5530","name":"ITMO","fullname":"ITMO University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/fBvTjWt5kZM802VUpqCno.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63a041c56b087d7413be69e8","avatarUrl":"/avatars/66593b5a09540f343377e8dffe9e0100.svg","isPro":false,"fullname":"Alexey Gavrilov","user":"aogavrilov","type":"user"},{"_id":"661d4fdfbca1a6038b362e8a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/AcSzMXgu7FaXn6FHvFAkA.jpeg","isPro":false,"fullname":"Nikolay Nikitin","user":"nnikitin","type":"user"},{"_id":"63654b82ad2c9e8b672d8a4b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667661623748-63654b82ad2c9e8b672d8a4b.jpeg","isPro":false,"fullname":"Alan Gazzaev","user":"AlanRobotics","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"61b8da8ed48e2b6a4d2f5530","name":"ITMO","fullname":"ITMO University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/fBvTjWt5kZM802VUpqCno.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.28443.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.28443 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.28443 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.28443 in a Space README.md to link it from this page.

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