Video Unlearning via Low-Rank Refusal Vector

• URL: http://arxiv.org/abs/2506.07891v1
• Date: Mon, 09 Jun 2025 16:06:49 GMT
• Title: Video Unlearning via Low-Rank Refusal Vector
• Authors: Simone Facchiano, Stefano Saravalle, Matteo Migliarini, Edoardo De Matteis, Alessio Sampieri, Andrea Pilzer, Emanuele Rodolà, Indro Spinelli, Luca Franco, Fabio Galasso,
• Abstract summary: Video generative models democratize the creation of visual content through intuitive instruction following.<n>But they also inherit the biases and harmful concepts embedded within their web-scale training data.<n>This inheritance creates a significant risk, as users can readily generate undesirable and even illegal content.<n>This work introduces the first unlearning technique tailored explicitly for video diffusion models to address this critical issue.
• Score: 16.820840077695912
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Video generative models democratize the creation of visual content through intuitive instruction following, but they also inherit the biases and harmful concepts embedded within their web-scale training data. This inheritance creates a significant risk, as users can readily generate undesirable and even illegal content. This work introduces the first unlearning technique tailored explicitly for video diffusion models to address this critical issue. Our method requires 5 multi-modal prompt pairs only. Each pair contains a "safe" and an "unsafe" example that differ only by the target concept. Averaging their per-layer latent differences produces a "refusal vector", which, once subtracted from the model parameters, neutralizes the unsafe concept. We introduce a novel low-rank factorization approach on the covariance difference of embeddings that yields robust refusal vectors. This isolates the target concept while minimizing collateral unlearning of other semantics, thus preserving the visual quality of the generated video. Our method preserves the model's generation quality while operating without retraining or access to the original training data. By embedding the refusal direction directly into the model's weights, the suppression mechanism becomes inherently more robust against adversarial bypass attempts compared to surface-level input-output filters. In a thorough qualitative and quantitative evaluation, we show that we can neutralize a variety of harmful contents, including explicit nudity, graphic violence, copyrights, and trademarks. Project page: https://www.pinlab.org/video-unlearning.

Related papers

• Embedding Hidden Adversarial Capabilities in Pre-Trained Diffusion Models [1.534667887016089]
  We introduce a new attack paradigm that embeds hidden adversarial capabilities directly into diffusion models via fine-tuning.<n>The resulting tampered model generates high-quality images indistinguishable from those of the original.<n>We demonstrate the effectiveness and stealthiness of our approach, uncovering a covert attack vector that raises new security concerns.
  arXiv  Detail & Related papers  (2025-04-05T12:51:36Z)
• Concept Steerers: Leveraging K-Sparse Autoencoders for Controllable Generations [10.86252546314626]
  Text-to-image generative models are prone to adversarial attacks and inadvertently generate unsafe, unethical content.<n>We propose a novel framework leveraging k-sparse autoencoders (k-SAEs) to enable efficient and interpretable concept manipulation.<n>Our method yields an improvement of $mathbf20.01%$ in unsafe concept removal, is effective in style manipulation, and is $mathbfsim5$x faster than current state-of-the-art.
  arXiv  Detail & Related papers  (2025-01-31T11:52:47Z)
• SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders [4.013156524547073]
  Diffusion models can inadvertently generate harmful or undesirable content.<n>Recent machine unlearning approaches offer potential solutions but often lack transparency.<n>We introduce SAeUron, a novel method leveraging features learned by sparse autoencoders to remove unwanted concepts.
  arXiv  Detail & Related papers  (2025-01-29T23:29:47Z)
• Score Forgetting Distillation: A Swift, Data-Free Method for Machine Unlearning in Diffusion Models [63.43422118066493]
  Machine unlearning (MU) is a crucial foundation for developing safe, secure, and trustworthy GenAI models.<n>Traditional MU methods often rely on stringent assumptions and require access to real data.<n>This paper introduces Score Forgetting Distillation (SFD), an innovative MU approach that promotes the forgetting of undesirable information in diffusion models.
  arXiv  Detail & Related papers  (2024-09-17T14:12:50Z)
• Probing Unlearned Diffusion Models: A Transferable Adversarial Attack Perspective [20.263233740360022]
  Unlearning methods have been developed to erase concepts from diffusion models. This paper aims to leverage the transferability of the adversarial attack to probe the unlearning robustness under a black-box setting. Specifically, we employ an adversarial search strategy to search for the adversarial embedding which can transfer across different unlearned models.
  arXiv  Detail & Related papers  (2024-04-30T09:14:54Z)
• REST: REtrieve & Self-Train for generative action recognition [54.90704746573636]
  We propose to adapt a pre-trained generative Vision & Language (V&L) Foundation Model for video/action recognition. We show that direct fine-tuning of a generative model to produce action classes suffers from severe overfitting. We introduce REST, a training framework consisting of two key components.
  arXiv  Detail & Related papers  (2022-09-29T17:57:01Z)
• MOVE: Effective and Harmless Ownership Verification via Embedded External Features [104.97541464349581]
  We propose an effective and harmless model ownership verification (MOVE) to defend against different types of model stealing simultaneously.<n>We conduct the ownership verification by verifying whether a suspicious model contains the knowledge of defender-specified external features.<n>We then train a meta-classifier to determine whether a model is stolen from the victim.
  arXiv  Detail & Related papers  (2022-08-04T02:22:29Z)
• Diffusion Models for Adversarial Purification [69.1882221038846]
  Adrial purification refers to a class of defense methods that remove adversarial perturbations using a generative model. We propose DiffPure that uses diffusion models for adversarial purification. Our method achieves the state-of-the-art results, outperforming current adversarial training and adversarial purification methods.
  arXiv  Detail & Related papers  (2022-05-16T06:03:00Z)
• Stylized Adversarial Defense [105.88250594033053]
  adversarial training creates perturbation patterns and includes them in the training set to robustify the model. We propose to exploit additional information from the feature space to craft stronger adversaries. Our adversarial training approach demonstrates strong robustness compared to state-of-the-art defenses.
  arXiv  Detail & Related papers  (2020-07-29T08:38:10Z)
• TransMoMo: Invariance-Driven Unsupervised Video Motion Retargeting [107.39743751292028]
  TransMoMo is capable of transferring motion of a person in a source video realistically to another video of a target person. We exploit invariance properties of three factors of variation including motion, structure, and view-angle. We demonstrate the effectiveness of our method over the state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-03-31T17:49:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.