Lune

CVPR Findings2026

Now You See It, Now You Don't: Instant Concept Erasure for Safe Text-to-Image and Video Generation

Shristi Das Biswas, Arani Roy, Kaushik Roy

Abstract

Robust concept removal for text-to-image (T2I) and textto-video (T2V) models is essential for their safe deployment. Existing methods, however, suffer from costly retraining, inference overhead, or vulnerability to adversarial attacks. Crucially, they rarely model the latent semantic overlap between the target erase concept and surrounding content -causing collateral damage post-erasure -and even fewer methods work reliably across both T2I and T2V domains. We introduce Instant Concept Erasure (ICE), a training-free, modality-agnostic, one-shot weight modification approach that achieves precise, persistent unlearning with zero overhead. ICE defines erase and preserve subspaces using anisotropic energy-weighted scaling, then explicitly regularizes against their intersection using a unique, closed-form overlap projector. We pose a convex and Lipschitz-bounded Spectral Unlearning Objective, balancing erasure fidelity and intersection preservation, that admits a stable and unique analytical solution. This solution defines a dissociation operator that is translated to the model's text-conditioning layers, making the edit permanent and runtime-free. Across targeted removals of artistic styles, objects, identities, and explicit content, ICE efficiently achieves strong erasure with improved robustness to red-teaming, all while causing only minimal degradation of original generative abilities in both T2I and T2V models.

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.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models cites Learning Transferable Visual Models From Natural Language Supervision
  • Directly Denoising Diffusion Models cites High-Resolution Image Synthesis with Latent Diffusion Models
  • Directly Denoising Diffusion Models cites GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
  • Directly Denoising Diffusion Models cites Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
  • Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding cites High-Resolution Image Synthesis with Latent Diffusion Models
  • Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding cites GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
  • Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding cites Learning Transferable Visual Models From Natural Language Supervision
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on37

  1. Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
  2. High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser et al.CVPR 2022 · 13,123 citations
  3. Directly Denoising Diffusion ModelsDan Zhang, Jingjing Wang, Feng LuoICML 2024 · 11,724 citations
  4. Photorealistic Text-to-Image Diffusion Models with Deep Language UnderstandingChitwan Saharia, William Chan, Saurabh Saxena, Lala Li et al.NeurIPS 2022 · 8,965 citations
  5. GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion ModelsAlexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam et al.ICML 2022 · 4,691 citations

Related papers

  1. ICE: Intercede Concept Erasure in Text-to-Image Diffusion ModelsYizhou Lin, Nisha Huang, Kaer Huang, Henglin Liu et al.ACM MM 2025 · 1 citation
  2. CURE: Concept Unlearning via Orthogonal Representation Editing in Diffusion ModelsShristi Das Biswas, Arani Roy, Kaushik RoyNeurIPS 2025 · 30 citations
  3. Beyond Text Prompts: Precise Concept Erasure through Text-Image CollaborationJun Li, Lizhi Xiong, Ziqiang Li, Weiwei Jiang et al.CVPR 2026 · 1 citation
  4. ForceForget: Reinforcement Concept Removal for Enhancing Safety in Text-to-Image ModelsDong Han, Yong LiICML 2026
  5. Closed-Form Concept Erasure via Double ProjectionsChi Zhang, Jingpu Cheng, Zhixian Wang, Ping LiuCVPR 2026 · 5 citations
Dusk over the sea between two cliffs drawn in fine vertical lines