Semantic Surgery: Zero-Shot Concept Erasure in Diffusion Models

TL;DR

Semantic Surgery is a training-free, zero-shot method for concept erasure in diffusion models that improves completeness and locality without sacrificing image quality, enhancing safety in text-to-image generation.

Contribution

It introduces a novel, training-free framework that performs dynamic, zero-shot concept erasure directly on text embeddings, outperforming existing methods in robustness and quality.

Findings

Achieves 93.58 H-score in object erasure

Reduces explicit content to just 1 instance

Attains 8.09 H_a in style erasure without quality loss

Abstract

Concept erasure in text-to-image diffusion models is crucial for mitigating harmful content, yet existing methods often compromise generative quality. We introduce Semantic Surgery, a novel training-free, zero-shot framework for concept erasure that operates directly on text embeddings before the diffusion process. It dynamically estimates the presence of target concepts in a prompt and performs a calibrated vector subtraction to neutralize their influence at the source, enhancing both erasure completeness and locality. The framework includes a Co-Occurrence Encoding module for robust multi-concept erasure and a visual feedback loop to address latent concept persistence. As a training-free method, Semantic Surgery adapts dynamically to each prompt, ensuring precise interventions. Extensive experiments on object, explicit content, artistic style, and multi-celebrity erasure tasks show our method significantly outperforms state-of-the-art approaches. We achieve superior completeness and robustness while preserving locality and image quality (e.g., 93.58 H-score in object erasure, reducing explicit content to just 1 instance, and 8.09 H_a in style erasure with no quality degradation). This robustness also allows our framework to function as a built-in threat detection system, offering a practical solution for safer text-to-image generation.