<video src=\"https://cdn-uploads.huggingface.co/production/uploads/62e18206926f4892a4c782bd/oa9OkTR6k7JtEX-W_zgsi.mp4\" controls=\"\" class=\"max-w-full!\"></video></p>\n","updatedAt":"2026-08-31T06:52:51.415Z","author":{"_id":"62e18206926f4892a4c782bd","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62e18206926f4892a4c782bd/m_OAeRb2a7a5PlnK6d4Dg.jpeg","fullname":"Chenguo Lin","name":"chenguolin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":11,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.4543130695819855},"editors":["chenguolin"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/62e18206926f4892a4c782bd/m_OAeRb2a7a5PlnK6d4Dg.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.26794","authors":[{"_id":"6a951882073195fee51572b5","name":"Bowen Xue","hidden":false},{"_id":"6a951882073195fee51572b6","name":"Brandon Y. Feng","hidden":false},{"_id":"6a951882073195fee51572b7","name":"Chenguo Lin","hidden":false},{"_id":"6a951882073195fee51572b8","name":"Yuchen Lin","hidden":false},{"_id":"6a951882073195fee51572b9","name":"Yujia Zeng","hidden":false},{"_id":"6a951882073195fee51572ba","name":"Lvmin Zhang","hidden":false},{"_id":"6a951882073195fee51572bb","name":"Maneesh Agrawala","hidden":false},{"_id":"6a951882073195fee51572bc","name":"Honglei Yan","hidden":false},{"_id":"6a951882073195fee51572bd","name":"Panwang Pan","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/62e18206926f4892a4c782bd/WbCfwvUWn_WZ3s3W9iChx.mp4"],"publishedAt":"2026-08-27T00:00:00.000Z","submittedOnDailyAt":"2026-08-31T00:00:00.000Z","title":"Ring Forcing: Towards Precise Long-Term Memory for Autoregressive Video Diffusion","submittedOnDailyBy":{"_id":"62e18206926f4892a4c782bd","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62e18206926f4892a4c782bd/m_OAeRb2a7a5PlnK6d4Dg.jpeg","isPro":false,"fullname":"Chenguo Lin","user":"chenguolin","type":"user","name":"chenguolin"},"summary":"Scaling video generation to long durations reveals a critical bottleneck: current models lack robust long-term memory. This deficiency can be studied along two critical aspects: object permanence, the ability to precisely reproduce the appearance of objects upon re-entry; and memory capacity, the ability to process ultra-long context and use information from distant history. Robust long-term memory requires both: object permanence without sufficient context handling limits the temporal scope, while long context length without permanence fails to maintain identity. To address this, we present Ring Forcing, an autoregressive video diffusion framework designed to robustly construct and precisely utilize long-term memory. Our ring-structured training strategy enforces retrieval from distant history, effectively reconciling the trade-off between strict historical adherence and generative diversity. To expand memory capacity, we introduce a compression and timestep composition strategy. Under fixed sequence length constraints, this method extends the effective historical span to minutes-long durations and achieves a comprehensive receptive field over the entire history. Furthermore, we present a sparse RoPE mechanism to enable flexible, scalable memory adaptation while fully exploiting pre-trained priors. Extensive experiments demonstrate that Ring Forcing achieves superior minutes-long coherence and object permanence, significantly outperforming state-of-the-art methods.","upvotes":5,"discussionId":"6a951882073195fee51572be","projectPage":"https://ringforcing.com","ai_summary":"Ring Forcing is an autoregressive video diffusion framework that improves long-term memory through ring-structured training, history compression, and sparse rotary embeddings to achieve minute-long coherence.","ai_keywords":["autoregressive video diffusion","object permanence","memory capacity","Ring Forcing","ring-structured training","compression and timestep composition","sparse RoPE","rotary position embedding"],"ai_summary_model":"thinkingmachines/Inkling-Small"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"62e18206926f4892a4c782bd","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62e18206926f4892a4c782bd/m_OAeRb2a7a5PlnK6d4Dg.jpeg","isPro":false,"fullname":"Chenguo Lin","user":"chenguolin","type":"user"},{"_id":"6899c1c4c8d32e4cdea87215","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/8plwumYYJa1TNkZvF1xoj.jpeg","isPro":false,"fullname":"Bowen Xue","user":"BowenXue","type":"user"},{"_id":"678f3bc276152df1fa2563d1","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/lxPXcJyVJ-33yWtMN8Iuk.png","isPro":false,"fullname":"Ivystream","user":"Ivystream","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":"65f30652b0d359b2ffa4a42c","avatarUrl":"/avatars/f6cb0705b25cc0ccc0b099811ef9a871.svg","isPro":false,"fullname":"zrx","user":"z-rx","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.26794.md","query":{}}">
Ring Forcing: Towards Precise Long-Term Memory for Autoregressive Video Diffusion
Abstract
Ring Forcing is an autoregressive video diffusion framework that improves long-term memory through ring-structured training, history compression, and sparse rotary embeddings to achieve minute-long coherence.
Scaling video generation to long durations reveals a critical bottleneck: current models lack robust long-term memory. This deficiency can be studied along two critical aspects: object permanence, the ability to precisely reproduce the appearance of objects upon re-entry; and memory capacity, the ability to process ultra-long context and use information from distant history. Robust long-term memory requires both: object permanence without sufficient context handling limits the temporal scope, while long context length without permanence fails to maintain identity. To address this, we present Ring Forcing, an autoregressive video diffusion framework designed to robustly construct and precisely utilize long-term memory. Our ring-structured training strategy enforces retrieval from distant history, effectively reconciling the trade-off between strict historical adherence and generative diversity. To expand memory capacity, we introduce a compression and timestep composition strategy. Under fixed sequence length constraints, this method extends the effective historical span to minutes-long durations and achieves a comprehensive receptive field over the entire history. Furthermore, we present a sparse RoPE mechanism to enable flexible, scalable memory adaptation while fully exploiting pre-trained priors. Extensive experiments demonstrate that Ring Forcing achieves superior minutes-long coherence and object permanence, significantly outperforming state-of-the-art methods.
Community
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/2608.26794 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.26794 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.26794 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.