Cover Image

Sealing The Backdoor: Unlearning Adversarial Text Triggers In Diffusion Models Using Knowledge Distillation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Aravindan, Ashwath Vaithinathan, Jha, Abha, Salaway, Matthew, Bhide, Atharva Sandeep, Yaldiz, Duygu Nur
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Backdoor Defense in Diffusion Models via Spatial Attention Unlearning
by: Jha, Abha, et al.
Published: (2025)

Do VLMs Have Bad Eyes? Diagnosing Compositional Failures via Mechanistic Interpretability
by: Aravindan, Ashwath Vaithinathan, et al.
Published: (2025)

Second Guess: Detecting Uncertainty Through Abstention and Answer Stability in Small Language Models
by: Aravindan, Ashwath Vaithinathan, et al.
Published: (2026)

Fragile Thoughts: How Large Language Models Handle Chain-of-Thought Perturbations
by: Aravindan, Ashwath Vaithinathan, et al.
Published: (2026)

Code-Driven Planning in Grid Worlds with Large Language Models
by: Aravindan, Ashwath Vaithinathan, et al.
Published: (2025)

Rewarding Intellectual Humility Learning When Not To Answer In Large Language Models
by: Jha, Abha, et al.
Published: (2026)

BackWeak: Backdooring Knowledge Distillation Simply with Weak Triggers and Fine-tuning
by: Wang, Shanmin, et al.
Published: (2025)

CroMo-Mixup: Augmenting Cross-Model Representations for Continual Self-Supervised Learning
by: Mushtaq, Erum, et al.
Published: (2024)

Trigger without Trace: Towards Stealthy Backdoor Attack on Text-to-Image Diffusion Models
by: Zhang, Jie, et al.
Published: (2025)

Removing the Trigger, Not the Backdoor: Alternative Triggers and Latent Backdoors
by: Abad, Gorka, et al.
Published: (2026)

Unlearning Backdoor Threats: Enhancing Backdoor Defense in Multimodal Contrastive Learning via Local Token Unlearning
by: Liang, Siyuan, et al.
Published: (2024)

Robust MLLM Unlearning via Visual Knowledge Distillation
by: Wang, Yuhang, et al.
Published: (2025)

Backdoor Unlearning by Linear Task Decomposition
by: Abdelraheem, Amel, et al.
Published: (2025)

Scaling Exposes the Trigger: Input-Level Backdoor Detection in Text-to-Image Diffusion Models via Cross-Attention Scaling
by: Li, Zida, et al.
Published: (2026)

Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learning
by: Ali, Hassan, et al.
Published: (2024)

Probing Unlearned Diffusion Models: A Transferable Adversarial Attack Perspective
by: Han, Xiaoxuan, et al.
Published: (2024)

Adv-KD: Adversarial Knowledge Distillation for Faster Diffusion Sampling
by: Mekonnen, Kidist Amde, et al.
Published: (2024)

TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models
by: Xiang, Qianlong, et al.
Published: (2026)

Adversarial Backdoor Defense in CLIP
by: Kuang, Junhao, et al.
Published: (2024)

Using Interleaved Ensemble Unlearning to Keep Backdoors at Bay for Finetuning Vision Transformers
by: Li, Zeyu Michael
Published: (2024)

Dynamic Guidance Adversarial Distillation with Enhanced Teacher Knowledge
by: Park, Hyejin, et al.
Published: (2024)

Teacher-Guided Student Self-Knowledge Distillation Using Diffusion Model
by: Wang, Yu, et al.
Published: (2026)

Is It Possible to Backdoor Face Forgery Detection with Natural Triggers?
by: Han, Xiaoxuan, et al.
Published: (2023)

Explainability-Driven Leaf Disease Classification Using Adversarial Training and Knowledge Distillation
by: Echim, Sebastian-Vasile, et al.
Published: (2023)

Adversarial Concept Distillation for One-Step Diffusion Personalization
by: Yang, Yixiong, et al.
Published: (2025)

Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models
by: Wang, Zhongqi, et al.
Published: (2025)

T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models
by: Wang, Zhongqi, et al.
Published: (2024)

Unlearning Concepts from Text-to-Video Diffusion Models
by: Liu, Shiqi, et al.
Published: (2024)

Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models
by: Zhang, Yimeng, et al.
Published: (2024)

Holistic Unlearning Benchmark: A Multi-Faceted Evaluation for Text-to-Image Diffusion Model Unlearning
by: Moon, Saemi, et al.
Published: (2024)

Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
by: Lin, Weilin, et al.
Published: (2024)

Data-free Knowledge Distillation with Diffusion Models
by: Qi, Xiaohua, et al.
Published: (2025)

Erasure or Erosion? Evaluating Compositional Degradation in Unlearned Text-To-Image Diffusion Models
by: Koma, Arian Komaei, et al.
Published: (2026)

Backdoor Attack with Sparse and Invisible Trigger
by: Gao, Yinghua, et al.
Published: (2023)

Towards Adversarial Robustness And Backdoor Mitigation in SSL
by: Satpathy, Aryan, et al.
Published: (2024)

Seal2Real: Prompt Prior Learning on Diffusion Model for Unsupervised Document Seal Data Generation and Realisation
by: Yan, Mingfu, et al.
Published: (2023)

BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP
by: Bai, Jiawang, et al.
Published: (2023)

Twin Trigger Generative Networks for Backdoor Attacks against Object Detection
by: Li, Zhiying, et al.
Published: (2024)

Invisible Backdoor Triggers in Image Editing Model via Deep Watermarking
by: Chen, Yu-Feng, et al.
Published: (2025)

FVGen: Accelerating Novel-View Synthesis with Adversarial Video Diffusion Distillation
by: Teng, Wenbin, et al.
Published: (2025)