This approach first reframes concept erasure as prompt-conditioned parameter amortization and trains a hypernetwork to map textual descriptions to prompt-specific LoRA updates, eliminating the need for per-prompt gradient optimization or manual LoRA merging.
Abstract
Recent advances in text-to-image (T2I) generation have substantially improved visual synthesis, but have also raised increasing safety concerns due to their potential to generate harmful or undesirable content. Existing concept erasure methods predominantly follow a static weight paradigm, producing a single frozen adapter that struggles to adapt to diverse prompt variations and suffers from parameter interference when scaling to multiple concepts. We propose \textbf{HyperErase}, a framework for concept erasure based on hypernetwork-driven prompt-conditioned parameter synthesis. Our approach first reframes concept erasure as prompt-conditioned parameter amortization and trains a hypernetwork to map textual descriptions to prompt-specific LoRA updates, eliminating the need for per-prompt gradient optimization or manual LoRA merging. To further improve the stability and precision of synthesized adapters, we develop a decoupled rectification strategy, which disentangles LoRA tokens into pattern and scale subspaces, applies a square-root transform to curb multiplicative over-scaling, and leverages teacher-derived canonical priors for inference-time correction. Extensive experiments across major concept categories demonstrate that HyperErase consistently improves the trade-off between erasure effectiveness, image quality, and semantic alignment, achieving performance comparable to gold-standard single-concept baselines. Furthermore, the resulting models can provide specialized LoRAs for each input prompt variation in a single forward pass without requiring gradient updates during inference. These principled and flexible framework offers a new paradigm for concept erasure in T2I models.
This work proposes to erase concepts by directly manipulating the model's internal representations by operating exclusively on the sparse text-branch tokens and leveraging the straight sampling trajectory of rectified flow, achieving effective concept erasure with negligible overhead and without any training.
Qiao Li, Xiaomeng Fu, Yuanshu Zhao et al.· 1 citation
V-Engram is introduced, a trigger-indexed external memory mechanism for Stable Diffusion 3.5 that establishes trigger-indexed memory as a modular interface for adding targeted visual evidence without rewriting the generator.
Hao-Ran He, Run-Yuan Cai, Yi-Ming Wang et al.· 0 citations
STAGE confines each edit to the low-dimensional subspace spanned by the differences between erase and anchor embeddings, and relaxes the norm-preserving (orthogonal) constraint of prior editors into a least-squares affine correction that maps target activations onto safe anchors subject to a penalty on the displacement...
Karol Dziekan, P. Spurek, Dawid Malarz· 0 citations
A lightweight Text Encoder Alignment framework that fine-tunes only the text encoder while keeping the generative backbone fully frozen, and achieves state-of-the-art erasure robustness against black-box and white-box adversarial attacks on Stable Diffusion v1.4, while preserving generation quality on benign prompts.
This work proposes EraseSAE, a novel framework that leverages sparse autoencoders to achieve surgical concept erasure in DiT-based T2V diffusion models via a principled decompose-attribute-erase pipeline, and introduces the Partitioned Convolutional Sparse Autoencoder.
Xing-Hao Wang, Dong Li, Wei Yu et al.· 0 citations
This work proposes Hybrid-Policy Self-Distillation (HPSD), a novel self-distillation framework where a single TI2V model acts as both teacher and student under different conditions: the teacher operates in TI2V mode with a high-quality first frame and an enhanced prompt, while the student runs in the base T2V mode with...
Jia-Zi Bu, Peng-Yang Ling, Yu-Jie Zhou et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.