DecMem: Towards Minute-Long Consistent World Generation with Decoupled Memory</p>\n","updatedAt":"2026-06-01T09:37:24.887Z","author":{"_id":"661a59ff8858a270e6ad4481","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XZ0X-7HCjaw_0PpYM3Pz.png","fullname":"Zhenhao Yang","name":"JeffreyYzh","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7063995003700256},"editors":["JeffreyYzh"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XZ0X-7HCjaw_0PpYM3Pz.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.31336","authors":[{"_id":"6a1d297a808ddbc3c7d4367e","user":{"_id":"661a59ff8858a270e6ad4481","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XZ0X-7HCjaw_0PpYM3Pz.png","isPro":false,"fullname":"Zhenhao Yang","user":"JeffreyYzh","type":"user","name":"JeffreyYzh"},"name":"Zhenhao Yang","status":"claimed_verified","statusLastChangedAt":"2026-06-01T09:32:06.774Z","hidden":false},{"_id":"6a1d297a808ddbc3c7d4367f","name":"Xiaoshi Wu","hidden":false},{"_id":"6a1d297a808ddbc3c7d43680","name":"Zhengyao Lv","hidden":false},{"_id":"6a1d297a808ddbc3c7d43681","name":"Xiaoyu Shi","hidden":false},{"_id":"6a1d297a808ddbc3c7d43682","name":"Xintao Wang","hidden":false},{"_id":"6a1d297a808ddbc3c7d43683","name":"Pengfei Wan","hidden":false},{"_id":"6a1d297a808ddbc3c7d43684","name":"Kun Gai","hidden":false},{"_id":"6a1d297a808ddbc3c7d43685","name":"Kwan-Yee K. Wong","hidden":false}],"publishedAt":"2026-05-29T00:00:00.000Z","submittedOnDailyAt":"2026-06-01T00:00:00.000Z","title":"DecMem: Towards Minute-Long Consistent World Generation with Decoupled Memory","submittedOnDailyBy":{"_id":"661a59ff8858a270e6ad4481","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XZ0X-7HCjaw_0PpYM3Pz.png","isPro":false,"fullname":"Zhenhao Yang","user":"JeffreyYzh","type":"user","name":"JeffreyYzh"},"summary":"Recent advances in video generative models have promoted rapid progress in controllable world models. However, maintaining fine-grained spatio-temporal consistency under long-horizon reasoning remains a key challenge. In this work, we move beyond explicit 3D memory and coarse frame-level implicit modeling, and propose a fine-grained, learnable, and scalable memory for consistent world generation. We first identify two fundamental limitations of naïve learnable memory architectures in long-horizon extrapolation, namely computational inefficiency and attention dispersion. Through a systematic analysis of attention dispersion, we propose DecMem, a decoupled memory architecture that employs Sparse Global Memory for efficient fine-grained access to global history and Anchored Local Memory for stable and high-quality extrapolation. Extensive experiments demonstrate that DecMem significantly outperforms current state-of-the-art methods. By ensuring precise and efficient long-term memory and achieving superior extrapolation capabilities, DecMem enables minute-level controllable long video generation with high fidelity and consistency.","upvotes":3,"discussionId":"6a1d297a808ddbc3c7d43686","projectPage":"https://jeffreyyzh.github.io/DecMem-Page/","githubRepo":"https://github.com/KlingAIResearch/DecMem","githubRepoAddedBy":"user","ai_summary":"A novel decoupled memory architecture called DecMem is introduced for consistent long-horizon video generation, addressing computational inefficiency and attention dispersion issues in learnable memory systems.","ai_keywords":["video generative models","world models","spatio-temporal consistency","long-horizon reasoning","learnable memory","attention dispersion","sparse global memory","anchored local memory","video generation","extrapolation"],"githubStars":3,"organization":{"_id":"662c559b322afcbae51b3c8b","name":"KlingTeam","fullname":"Kling Team","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e272ca6c78a8c122b12127/ZQV1aKLUDPf2rUcxxAqj6.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"661a59ff8858a270e6ad4481","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XZ0X-7HCjaw_0PpYM3Pz.png","isPro":false,"fullname":"Zhenhao Yang","user":"JeffreyYzh","type":"user"},{"_id":"662f93942510ef5735d7ad00","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/662f93942510ef5735d7ad00/ZIDIPm63sncIHFTT5b0uR.png","isPro":false,"fullname":"magicwpf","user":"magicwpf","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"662c559b322afcbae51b3c8b","name":"KlingTeam","fullname":"Kling Team","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e272ca6c78a8c122b12127/ZQV1aKLUDPf2rUcxxAqj6.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.31336.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2605.31336 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2605.31336 in a Space README.md to link it from this page.

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