{"ID":2831659,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.07469","arxiv_id":"2512.07469","title":"VideoCoF: Unified Video Editing with Temporal Reasoner","abstract":"Existing video editing methods face a critical trade-off: expert models offer precision but rely on task-specific priors like masks, hindering unification; conversely, unified temporal in-context learning models are mask-free but lack explicit spatial cues, leading to weak instruction-to-region mapping and imprecise localization. To resolve this conflict, we propose VideoCoF, a novel Chain-of-Frames approach inspired by Chain-of-Thought reasoning. VideoCoF enforces a ``see, reason, then edit\" procedure by compelling the video diffusion model to first predict reasoning tokens (edit-region latents) before generating the target video tokens. This explicit reasoning step removes the need for user-provided masks while achieving precise instruction-to-region alignment and fine-grained video editing. Furthermore, we introduce a RoPE alignment strategy that leverages these reasoning tokens to ensure motion alignment and enable length extrapolation beyond the training duration. We demonstrate that with a minimal data cost of only 50k video pairs, VideoCoF achieves state-of-the-art performance on VideoCoF-Bench, validating the efficiency and effectiveness of our approach. Our code, weight, data are available at https://github.com/knightyxp/VideoCoF.","short_abstract":"Existing video editing methods face a critical trade-off: expert models offer precision but rely on task-specific priors like masks, hindering unification; conversely, unified temporal in-context learning models are mask-free but lack explicit spatial cues, leading to weak instruction-to-region mapping and imprecise lo...","url_abs":"https://arxiv.org/abs/2512.07469","url_pdf":"https://arxiv.org/pdf/2512.07469v2","authors":"[\"Xiangpeng Yang\",\"Ji Xie\",\"Yiyuan Yang\",\"Yue Ma\",\"Yan Huang\",\"Min Xu\",\"Qiang Wu\"]","published":"2025-12-08T11:50:18Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":606148,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2831659,"paper_url":"https://arxiv.org/abs/2512.07469","paper_title":"VideoCoF: Unified Video Editing with Temporal Reasoner","repo_url":"https://github.com/knightyxp/VideoCoF","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}