Lune

CVPR2025Top-tier venue

Unity in Diversity: Video Editing via Gradient-Latent Purification

Junyu Gao, Kunlin Yang, Xuan Yao, Yufan Hu

Abstract

Recently, text-driven video editing methods that optimize target latent representations have garnered significant attention and demonstrated promising results. However, these methods rely on self-supervised objectives to compute the gradients needed for updating latent representations, which inevitably introduces gradient noise, compromising content generation quality. Additionally, it is challenging to determine the optimal stopping point for the editing process, making it difficult to achieve an optimal solution for the latent representation. To address these issues, we propose a unified gradient-latent purification framework that collects gradient and latent information across different stages to identify effective and concordant update directions. We design a local coordinate system construction method based on feature decomposition, enabling short-term gradients and final-stage latents to be reprojected onto new axes. Then, we employ tailored coefficient regularization terms to effectively aggregate the decomposed information. Additionally, a temporal smoothing axis extension strategy is developed to enhance the temporal coherence of the generated content. Extensive experiments demonstrate that our proposed method outperforms state-of-the-art methods across various editing tasks, delivering superior editing performance. Project page is available in https://unity- in-diversity-editing.github.io.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

4 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Scalable Diffusion Models with Transformers cites Denoising Diffusion Probabilistic Models
  • Scalable Diffusion Models with Transformers cites Denoising Diffusion Implicit Models
  • LIVEditor-14B: Lightning Unified Video Editing via In-Context Sparse Attention cites Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset
  • Denoising Diffusion Implicit Models cites Denoising Diffusion Probabilistic Models
  • Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset cites Scalable Diffusion Models with Transformers
  • FIA-Edit: Frequency-Interactive Attention for Efficient and High-Fidelity Inversion-Free Text-Guided Image Editing cites Scalable Diffusion Models with Transformers
  • FIA-Edit: Frequency-Interactive Attention for Efficient and High-Fidelity Inversion-Free Text-Guided Image Editing cites Denoising Diffusion Implicit Models
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers4

  1. Scaling Instruction-Based Video Editing with a High-Quality Synthetic DatasetQingyan Bai, Qiuyu Wang, Hao Ouyang, Yue Yu et al.CVPR 2026 · 79 citations
  2. FIA-Edit: Frequency-Interactive Attention for Efficient and High-Fidelity Inversion-Free Text-Guided Image EditingKaixiang Yang, Boyang Shen, Xin Li, Yuchen Dai et al.AAAI 2026 · 3 citations
  3. LIVEditor-14B: Lightning Unified Video Editing via In-Context Sparse AttentionShitong Shao, zikai ZHOU, Haopeng Li, Yingwei Song et al.ICML 2026 · 2 citations
  4. SC2^{2}-WM: A Self-Correcting World Model with Closed-Loop Feedback for Vision-and-Language Navigation in Continuous EnvironmentsXuan Yao, Yuze Zhu, JUNYU GAO, Zongmeng Wang et al.ICML 2026

Builds on42

  1. Denoising Diffusion Probabilistic ModelsJonathan Ho, Ajay Jain, Pieter AbbeelNeurIPS 2020 · 35,902 citations
  2. Denoising Diffusion Implicit ModelsJiaming Song, Chenlin Meng, Stefano ErmonICLR 2021 · 11,743 citations
  3. Emerging Properties in Self-Supervised Vision TransformersMathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou et al.ICCV 2021 · 8,921 citations
  4. Scalable Diffusion Models with TransformersWilliam Peebles, Saining XieICCV 2023 · 5,568 citations
  5. Gradient Surgery for Multi-Task LearningTianhe Yu, Saurabh Kumar, Abhishek Gupta, Sergey Levine et al.NeurIPS 2020 · 2,261 citations

Related papers

  1. SUV: Suppressing Undesired Video Content via Semantic Modulation Based on Text EmbeddingsXiang Lv, Mingwen Shao, Lingzhuang Meng, Chang Liu et al.ICCV 2025 · 2 citations
  2. Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-ResolutionShangchen Zhou, Peiqing Yang, Jianyi Wang, Yihang Luo et al.CVPR 2024 · 52 citations
  3. UniEdit: A Unified Tuning-Free Framework for Video Motion and Appearance EditingJianhong Bai, Tianyu He, Yuchi Wang, Junliang Guo et al.ACM MM 2025 · 6 citations
  4. StableVideo: Text-driven Consistency-aware Diffusion Video EditingWenhao Chai, Xun Guo, Gaoang Wang, Yan LuICCV 2023 · 219 citations
  5. Visual Prompting for One-shot Controllable Video Editing without InversionZhengbo Zhang, Yuxi Zhou, Duo Peng, Joo-Hwee Lim et al.CVPR 2025
Dusk over the sea between two cliffs drawn in fine vertical lines