Backdoor Defense in Diffusion Models via Spatial Attention Unlearning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jha, Abha, Aravindan, Ashwath Vaithinathan, Salaway, Matthew, Bhide, Atharva Sandeep, Yaldiz, Duygu Nur
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910919260372992
author Jha, Abha
Aravindan, Ashwath Vaithinathan
Salaway, Matthew
Bhide, Atharva Sandeep
Yaldiz, Duygu Nur
author_facet Jha, Abha
Aravindan, Ashwath Vaithinathan
Salaway, Matthew
Bhide, Atharva Sandeep
Yaldiz, Duygu Nur
contents Text-to-image diffusion models are increasingly vulnerable to backdoor attacks, where malicious modifications to the training data cause the model to generate unintended outputs when specific triggers are present. While classification models have seen extensive development of defense mechanisms, generative models remain largely unprotected due to their high-dimensional output space, which complicates the detection and mitigation of subtle perturbations. Defense strategies for diffusion models, in particular, remain under-explored. In this work, we propose Spatial Attention Unlearning (SAU), a novel technique for mitigating backdoor attacks in diffusion models. SAU leverages latent space manipulation and spatial attention mechanisms to isolate and remove the latent representation of backdoor triggers, ensuring precise and efficient removal of malicious effects. We evaluate SAU across various types of backdoor attacks, including pixel-based and style-based triggers, and demonstrate its effectiveness in achieving 100% trigger removal accuracy. Furthermore, SAU achieves a CLIP score of 0.7023, outperforming existing methods while preserving the model's ability to generate high-quality, semantically aligned images. Our results show that SAU is a robust, scalable, and practical solution for securing text-to-image diffusion models against backdoor attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18563
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Backdoor Defense in Diffusion Models via Spatial Attention Unlearning
Jha, Abha
Aravindan, Ashwath Vaithinathan
Salaway, Matthew
Bhide, Atharva Sandeep
Yaldiz, Duygu Nur
Cryptography and Security
Artificial Intelligence
Computer Vision and Pattern Recognition
Text-to-image diffusion models are increasingly vulnerable to backdoor attacks, where malicious modifications to the training data cause the model to generate unintended outputs when specific triggers are present. While classification models have seen extensive development of defense mechanisms, generative models remain largely unprotected due to their high-dimensional output space, which complicates the detection and mitigation of subtle perturbations. Defense strategies for diffusion models, in particular, remain under-explored. In this work, we propose Spatial Attention Unlearning (SAU), a novel technique for mitigating backdoor attacks in diffusion models. SAU leverages latent space manipulation and spatial attention mechanisms to isolate and remove the latent representation of backdoor triggers, ensuring precise and efficient removal of malicious effects. We evaluate SAU across various types of backdoor attacks, including pixel-based and style-based triggers, and demonstrate its effectiveness in achieving 100% trigger removal accuracy. Furthermore, SAU achieves a CLIP score of 0.7023, outperforming existing methods while preserving the model's ability to generate high-quality, semantically aligned images. Our results show that SAU is a robust, scalable, and practical solution for securing text-to-image diffusion models against backdoor attacks.
title Backdoor Defense in Diffusion Models via Spatial Attention Unlearning
topic Cryptography and Security
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.18563