<a href=\"https://cdn-uploads.huggingface.co/production/uploads/67f87529318a17cc80365190/mOm1eNwWKUtkCNixbsukM.png\" rel=\"nofollow\"><img src=\"https://cdn-uploads.huggingface.co/production/uploads/67f87529318a17cc80365190/mOm1eNwWKUtkCNixbsukM.png\" alt=\"image\"></a></p>\n","updatedAt":"2026-07-17T16:45:42.583Z","author":{"_id":"67f87529318a17cc80365190","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/67f87529318a17cc80365190/kv4cAvD5BrWQFRXKG4FXg.jpeg","fullname":"Maijunxian Wang","name":"Mark7121983123","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.39387303590774536},"editors":["Mark7121983123"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/67f87529318a17cc80365190/kv4cAvD5BrWQFRXKG4FXg.jpeg"],"reactions":[],"isReport":false}},{"id":"6a5afa00a1a4165c04edb25a","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":376,"isUserFollowing":false},"createdAt":"2026-07-18T03:58:56.000Z","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* [Next Forcing: Causal World Modeling with Multi-Chunk Prediction](https://huggingface.co/papers/2606.11187) (2026)\n* [Flex-Forcing: Towards a Unified Autoregressive and Bidirectional Video Diffusion Model](https://huggingface.co/papers/2607.03509) (2026)\n* [BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression](https://huggingface.co/papers/2606.10135) (2026)\n* [Latent Visual Cache for Video Reasoning](https://huggingface.co/papers/2607.02607) (2026)\n* [UNIVERSE: Unified Video Action Models for Autonomous Driving with Flexible Mask-Modulated Modality Generation](https://huggingface.co/papers/2607.05133) (2026)\n* [MoWorld: A Flash World Model](https://huggingface.co/papers/2607.06216) (2026)\n* [Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoning](https://huggingface.co/papers/2606.02842) (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/2606.11187\">Next Forcing: Causal World Modeling with Multi-Chunk Prediction</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2607.03509\">Flex-Forcing: Towards a Unified Autoregressive and Bidirectional Video Diffusion Model</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.10135\">BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2607.02607\">Latent Visual Cache for Video Reasoning</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2607.05133\">UNIVERSE: Unified Video Action Models for Autonomous Driving with Flexible Mask-Modulated Modality Generation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2607.06216\">MoWorld: A Flash World Model</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.02842\">Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoning</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>@librarian-bot recommend</code></p>\n","updatedAt":"2026-07-18T03:58:56.519Z","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":376,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7143896818161011},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.15278","authors":[{"_id":"6a5a5baad6689d7651ea876d","name":"Zezhong Qian","hidden":false},{"_id":"6a5a5baad6689d7651ea876e","name":"Xiaowei Chi","hidden":false},{"_id":"6a5a5baad6689d7651ea876f","name":"Chak-Wing Mak","hidden":false},{"_id":"6a5a5baad6689d7651ea8770","name":"Tianze Zhou","hidden":false},{"_id":"6a5a5baad6689d7651ea8771","name":"Ruibin Yuan","hidden":false},{"_id":"6a5a5baad6689d7651ea8772","name":"Yuhan Rui","hidden":false},{"_id":"6a5a5baad6689d7651ea8773","name":"Hengzhe Sun","hidden":false},{"_id":"6a5a5baad6689d7651ea8774","name":"Zhuoqun Wu","hidden":false},{"_id":"6a5a5baad6689d7651ea8775","name":"Yuming Li","hidden":false},{"_id":"6a5a5baad6689d7651ea8776","name":"Siyuan Qian","hidden":false},{"_id":"6a5a5baad6689d7651ea8777","name":"Sirui Han","hidden":false},{"_id":"6a5a5baad6689d7651ea8778","name":"Shanghang Zhang","hidden":false}],"publishedAt":"2026-07-16T00:00:00.000Z","submittedOnDailyAt":"2026-07-17T00:00:00.000Z","title":"Hierarchical Denoising For Multi-Step Visual Reasoning","submittedOnDailyBy":{"_id":"67f87529318a17cc80365190","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/67f87529318a17cc80365190/kv4cAvD5BrWQFRXKG4FXg.jpeg","isPro":false,"fullname":"Maijunxian Wang","user":"Mark7121983123","type":"user","name":"Mark7121983123"},"summary":"Video models are evolving into vision foundation models, yet they still lack human-like multi-step reasoning. Streaming autoregressive diffusion models are efficient but limited in reasoning, while bidirectional diffusion enables global revision with high inference costs due to dense frame-level denoising. Both paradigms struggle to achieve logical consistency and low-latency streaming for complex reasoning tasks. We propose HDR (Hierarchical Denoising for Visual Reasoning), a unified framework that integrates hierarchical latents into causal video generation for multi-step reasoning. HDR organizes video latents into a tree-structured hierarchy, enabling coarse-to-fine reasoning before streaming output. Coarse denoising layers preserve uncertain hypotheses for global planning, while finer layers progressively refine them into concrete visual states. A sparse hierarchical attention pattern (SHAP) further reduces temporal attention costs. We introduce a level-stratified multi-step video reasoning benchmark with out-of-distribution cases, covering six tasks: maze navigation, Tower of Hanoi, one-line drawing, sliding puzzle, Sokoban, and water pouring. Compared with streaming autoregressive diffusion baselines, HDR improves success from 34.22 to 60.29 (76.2% relative gain) and increases average progress from 76.00 to 89.56, demonstrating more consistent reasoning trajectories. HDR maintains low-latency streaming at 0.70 seconds per latent, achieving 54.2 times faster inference than bidirectional diffusion. It also retains 82.9% of full-data performance with only 2% training data, compared with 52.0% for bidirectional diffusion. Real-world robot experiments further demonstrate HDR's potential for physical interaction and world modeling. Project demo: https://hierarchical-diffusion-reasoning.github.io/.","upvotes":3,"discussionId":"6a5a5baad6689d7651ea8779","projectPage":"https://hierarchical-diffusion-reasoning.github.io/"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"},{"_id":"66bde8b33f0c44697d666f8f","avatarUrl":"/avatars/063f9727d80bf50a38d030866d99eb2d.svg","isPro":false,"fullname":"marco","user":"marcozzxx810","type":"user"},{"_id":"69ccaf5baa7e18f7efae4997","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Q9fpe3ZxK0-yt5QgZOd3U.jpeg","isPro":false,"fullname":"Мария Яковлев","user":"liamclar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.15278.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.15278 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.15278 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.15278 in a Space README.md to link it from this page.

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