<a href=\"https://cdn-uploads.huggingface.co/production/uploads/6554b73f1402f39ebc1bef37/Ca-8RxL1ArrvEnfGzvFBT.png\" rel=\"nofollow\"><img src=\"https://cdn-uploads.huggingface.co/production/uploads/6554b73f1402f39ebc1bef37/Ca-8RxL1ArrvEnfGzvFBT.png\" alt=\"微信图片_20260630180303_39959_120\"></a></p>\n","updatedAt":"2026-07-01T13:31:07.925Z","author":{"_id":"6554b73f1402f39ebc1bef37","avatarUrl":"/avatars/a79d6b06aff934b9b9e46271795feb78.svg","fullname":"Sen Liang","name":"bigfacing","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.3352333605289459},"editors":["bigfacing"],"editorAvatarUrls":["/avatars/a79d6b06aff934b9b9e46271795feb78.svg"],"reactions":[],"isReport":false}},{"id":"6a45c37e3b0fa8462bc933a7","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":372,"isUserFollowing":false},"createdAt":"2026-07-02T01:48:46.000Z","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* [CoVEBench: Can Video Editing Models Handle Complex Instructions?](https://huggingface.co/papers/2606.08415) (2026)\n* [JAVEDIT: Joint Audio-Visual Instruction-Guided Video Editing with Agentic Data Curation](https://huggingface.co/papers/2606.03168) (2026)\n* [TIDE: Task-Isolated Diffusion for Unified Video Editing and Generation](https://huggingface.co/papers/2606.08260) (2026)\n* [Sparkle: Realizing Lively Instruction-Guided Video Background Replacement via Decoupled Guidance](https://huggingface.co/papers/2605.06535) (2026)\n* [LoomVideo: Unifying Multimodal Inputs into Video Generation and Editing](https://huggingface.co/papers/2606.06042) (2026)\n* [EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis](https://huggingface.co/papers/2606.25465) (2026)\n* [TextSculptor: Training and Benchmarking Scene Text Editing](https://huggingface.co/papers/2605.21090) (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/2606.08415\">CoVEBench: Can Video Editing Models Handle Complex Instructions?</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.03168\">JAVEDIT: Joint Audio-Visual Instruction-Guided Video Editing with Agentic Data Curation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.08260\">TIDE: Task-Isolated Diffusion for Unified Video Editing and Generation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2605.06535\">Sparkle: Realizing Lively Instruction-Guided Video Background Replacement via Decoupled Guidance</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.06042\">LoomVideo: Unifying Multimodal Inputs into Video Generation and Editing</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2606.25465\">EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2605.21090\">TextSculptor: Training and Benchmarking Scene Text Editing</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":"2026-07-02T01:48:46.105Z","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":372,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7260882258415222},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.30599","authors":[{"_id":"6a45152a4f1dd35e48fb8cc1","name":"Sen Liang","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc2","name":"Cong Wang","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc3","name":"Zhentao Yu","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc4","name":"Fengbin Guan","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc5","name":"Zhengguang Zhou","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc6","name":"Teng Hu","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc7","name":"Youliang Zhang","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc8","name":"Yuan Zhou","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cc9","name":"Xin Li","hidden":false},{"_id":"6a45152a4f1dd35e48fb8cca","name":"Qinglin Lu","hidden":false},{"_id":"6a45152a4f1dd35e48fb8ccb","name":"Zhibo Chen","hidden":false}],"publishedAt":"2026-06-30T00:00:00.000Z","submittedOnDailyAt":"2026-07-01T00:00:00.000Z","title":"Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing","submittedOnDailyBy":{"_id":"6554b73f1402f39ebc1bef37","avatarUrl":"/avatars/a79d6b06aff934b9b9e46271795feb78.svg","isPro":false,"fullname":"Sen Liang","user":"bigfacing","type":"user","name":"bigfacing"},"summary":"Existing instruction-based video editing datasets commonly focus on single-task appearance editing, failing to meet the complex creative demands of real-world scenarios. To bridge this gap, we present Goku, a large-scale dataset featuring 2 million high-quality, instruction-aligned video editing pairs, which is the first to extend task boundaries from basic appearance editing to multi-task and structural manipulations(e.g., precise control of subject movement). To tackle the data synthesis challenges inherent in these complex tasks, we design an efficient data synthesis pipeline that decomposes complex edits into controllable sub-problems and introduce a progressive filtering system for data reliability throughout the whole process. Furthermore, we explore the optimal network structures on Goku, and propose Goku-Edit. To deeply comprehend complex editing instructions, Goku-Edit leverages an MLLM as its text encoder and adopts a decoupled dual-branch design: a dedicated mask branch handles structural control, freeing the main branch for appearance rendering. A comprehensive video editing benchmark, Goku-Bench, is also proposed with 1,000 human-verified test cases and 7 novel editing-specific metrics. Evaluated on Goku-Bench, Goku-Edit obtains up to +8% improvement on other open-source models in terms of instruction following.","upvotes":1,"discussionId":"6a45152a4f1dd35e48fb8ccc","projectPage":"https://flying-sky999.github.io/Goku.github.io/","ai_summary":"A large-scale video editing dataset and model are introduced that support multi-task and structural manipulations through advanced data synthesis and network architectures.","ai_keywords":["instruction-aligned video editing pairs","data synthesis pipeline","progressive filtering system","MLLM","decoupled dual-branch design","mask branch","appearance rendering","video editing benchmark","novel editing-specific metrics"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6554b73f1402f39ebc1bef37","avatarUrl":"/avatars/a79d6b06aff934b9b9e46271795feb78.svg","isPro":false,"fullname":"Sen Liang","user":"bigfacing","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.30599.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.30599 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.30599 in a Space README.md to link it from this page.

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