# Revoking Amnesia: RL-based Trajectory Optimization to Resurrect Erased Concepts in Diffusion Models

arXiv: https://arxiv.org/abs/2510.03302 · HTML: https://mlsanity.com/paper/2510.03302/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models.
- Builds on 38 of its 42 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Placed in

- Diffusion models / [Robustness, privacy and adversarial aspects](https://mlsanity.com/diffusion/robustness-privacy-and-adversarial-aspects/)

## Builds on (most cited first, 30 of 38)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [DanceGRPO: Unleashing GRPO on Visual Generation](https://mlsanity.com/paper/2505.07818/) (2025)
- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [Ablating Concepts in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.13516/) (2023)
- [MACE: Mass Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2403.06135/) (2024)
- [Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?](https://mlsanity.com/paper/2310.10012/) (2023)
- [To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now](https://mlsanity.com/paper/2310.11868/) (2023)
- [Large-scale Classification of Fine-Art Paintings: Learning The Right Metric on The Right Feature](https://mlsanity.com/paper/1505.00855/) (2015)
- [TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition](https://mlsanity.com/paper/2307.12493/) (2023)
- [Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts](https://mlsanity.com/paper/2309.06135/) (2023)
- [Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?](https://mlsanity.com/paper/2202.06675/) (2022)
- [Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to Advances](https://mlsanity.com/paper/2410.18775/) (2024)
- [Reinforcement Learning with Verifiable Rewards: GRPO's Effective Loss, Dynamics, and Success Amplification](https://mlsanity.com/paper/2503.06639/) (2025)
- [EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers](https://mlsanity.com/paper/2412.20413/) (2024)
- [Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Concepts](https://mlsanity.com/paper/2504.12782/) (2025)

## Built on by (most cited first, 11 of 11)

- [DragFlow: Unleashing DiT Priors with Region Based Supervision for Drag Editing](https://mlsanity.com/paper/2510.02253/) (2025)
- [Diffusion-Based Image Editing for Breaking Robust Watermarks](https://mlsanity.com/paper/2510.05978/) (2025)
- [Conditioned Activation Transport for T2I Safety Steering](https://mlsanity.com/paper/2603.03163/) (2026)
- [Vanishing Watermarks: Diffusion-Based Image Editing Undermines Robust Invisible Watermarking](https://mlsanity.com/paper/2602.20680/) (2026)
- [GraphiContact: Pose-aware Human-Scene Robust Contact Perception for Interactive Systems](https://mlsanity.com/paper/2603.20310/) (2026)
- [Editing on the Generative Manifold: A Theoretical and Empirical Study of General Diffusion-Based Image Editing Trade-offs](https://mlsanity.com/paper/2603.29736/) (2026)
- [GS-STVSR: Ultra-Efficient Continuous Spatio-Temporal Video Super-Resolution via 2D Gaussian Splatting](https://mlsanity.com/paper/2604.18047/) (2026)
- [DeRainMamba: A Frequency-Aware State Space Model with Detail Enhancement for Image Deraining](https://mlsanity.com/paper/2510.06746/) (2025)
- [When Denoising Becomes Unsigning: Theoretical and Empirical Analysis of Watermark Fragility Under Diffusion-Based Image Editing](https://mlsanity.com/paper/2603.04696/) (2026)
- [ReconMIL: Synergizing Latent Space Reconstruction with Bi-Stream Mamba for Whole Slide Image Analysis](https://mlsanity.com/paper/2603.19925/) (2026)
- [On the Controllability-Fidelity Frontier in Diffusion Editing](https://mlsanity.com/paper/2606.09901/) (2026)

## Abstract (arXiv)

Concept erasure techniques have been widely deployed in T2I diffusion models to prevent inappropriate content generation for safety and copyright considerations. However, as models evolve to next-generation architectures like Flux, established erasure methods (textite.g., ESD, UCE, AC) exhibit degraded effectiveness, raising questions about their true mechanisms. Through systematic analysis, we reveal that concept erasure creates only an illusion of ``amnesia": rather than genuine forgetting, these methods bias sampling trajectories away from target concepts, making the erasure fundamentally reversible. This insight motivates the need to distinguish superficial safety from genuine concept removal. In this work, we propose RevAm (underlineRevoking underlineAmnesia), an RL-based trajectory optimization framework that resurrects erased concepts by dynamically steering the denoising process without modifying model weights. By adapting Group Relative Policy Optimization (GRPO) to diffusion models, RevAm explores diverse recovery trajectories through trajectory-level rewards, overcoming local optima that limit existing methods. Extensive experiments demonstrate that RevAm achieves superior concept resurrection fidelity while reducing computational time by 10×, exposing critical vulnerabilities in current safety mechanisms and underscoring the need for more robust erasure techniques beyond trajectory manipulation.