Imagen de Portada

SRD: Reinforcement-Learned Semantic Perturbation for Backdoor Defense in VLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Xu, Shuhan, Liang, Siyuan, Zheng, Hongling, Liu, Aishan, Wang, Xinbiao, Luo, Yong, Lin, Fu, Rutkowski, Leszek, Tao, Dacheng
Formato:	Preprint
Publicado:	2025
Materias:	Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

CtrlAttack: A Unified Attack on World-Model Control in Diffusion Models
por: Xu, Shuhan, et al.
Publicado: (2026)

ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks
por: Ren, Zhiyao, et al.
Publicado: (2025)

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift
por: Liang, Siyuan, et al.
Publicado: (2024)

ME: Trigger Element Combination Backdoor Attack on Copyright Infringement
por: Yang, Feiyu, et al.
Publicado: (2025)

TrapFlow: Controllable Website Fingerprinting Defense via Dynamic Backdoor Learning
por: Liang, Siyuan, et al.
Publicado: (2024)

ELBA-Bench: An Efficient Learning Backdoor Attacks Benchmark for Large Language Models
por: Liu, Xuxu, et al.
Publicado: (2025)

Adversarial Backdoor Defense in CLIP
por: Kuang, Junhao, et al.
Publicado: (2024)

CleanerCLIP: Fine-grained Counterfactual Semantic Augmentation for Backdoor Defense in Contrastive Learning
por: Xun, Yuan, et al.
Publicado: (2024)

BadCLIP++: Stealthy and Persistent Backdoors in Multimodal Contrastive Learning
por: Liang, Siyuan, et al.
Publicado: (2026)

Towards Robust Physical-world Backdoor Attacks on Lane Detection
por: Zhang, Xinwei, et al.
Publicado: (2024)

Poisoned Forgery Face: Towards Backdoor Attacks on Face Forgery Detection
por: Liang, Jiawei, et al.
Publicado: (2024)

BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
por: Liang, Siyuan, et al.
Publicado: (2023)

VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models
por: Liang, Jiawei, et al.
Publicado: (2024)

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

Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack
por: Zhu, Mingli, et al.
Publicado: (2024)

Natural Reflection Backdoor Attack on Vision Language Model for Autonomous Driving
por: Liu, Ming, et al.
Publicado: (2025)

CogMorph: Cognitive Morphing Attacks for Text-to-Image Models
por: Jing, Zonglei, et al.
Publicado: (2025)

Who Generated This 3D Asset? Learning Source Attribution for Generative 3D Models
por: Ma, Sihan, et al.
Publicado: (2026)

Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt
por: Ying, Zonghao, et al.
Publicado: (2024)

Distillation Traps and Guards: A Calibration Knob for LLM Distillability
por: Zhan, Weixiao, et al.
Publicado: (2026)

MG-Net: Learn to Customize QAOA with Circuit Depth Awareness
por: Qian, Yang, et al.
Publicado: (2024)

Efficient Backdoor Defense in Multimodal Contrastive Learning: A Token-Level Unlearning Method for Mitigating Threats
por: Liu, Kuanrong, et al.
Publicado: (2024)

Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks
por: Ying, Zonghao, et al.
Publicado: (2024)

LanEvil: Benchmarking the Robustness of Lane Detection to Environmental Illusions
por: Zhang, Tianyuan, et al.
Publicado: (2024)

CopyrightShield: Enhancing Diffusion Model Security against Copyright Infringement Attacks
por: Guo, Zhixiang, et al.
Publicado: (2024)

TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting
por: Nguyen, Quang Duc, et al.
Publicado: (2026)

Review of Hallucination Understanding in Large Language and Vision Models
por: Ho, Zhengyi, et al.
Publicado: (2025)

Black-Box Adversarial Attack on Vision Language Models for Autonomous Driving
por: Wang, Lu, et al.
Publicado: (2025)

Object Detectors in the Open Environment: Challenges, Solutions, and Outlook
por: Liang, Siyuan, et al.
Publicado: (2024)

T2V-OptJail: Discrete Prompt Optimization for Text-to-Video Jailbreak Attacks
por: Liu, Jiayang, et al.
Publicado: (2025)

Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning
por: Ma, Guozheng, et al.
Publicado: (2025)

Multimodal Backdoor Attack on VLMs for Autonomous Driving via Graffiti and Cross-Lingual Triggers
por: Wang, Jiancheng, et al.
Publicado: (2026)

Transition Role of Entangled Data in Quantum Machine Learning
por: Wang, Xinbiao, et al.
Publicado: (2023)

Separable Power of Classical and Quantum Learning Protocols Through the Lens of No-Free-Lunch Theorem
por: Wang, Xinbiao, et al.
Publicado: (2024)

T2VShield: Model-Agnostic Jailbreak Defense for Text-to-Video Models
por: Liang, Siyuan, et al.
Publicado: (2025)

Lie Detector: Unified Backdoor Detection via Cross-Examination Framework
por: Wang, Xuan, et al.
Publicado: (2025)

Better, Faster: Harnessing Self-Improvement in Large Reasoning Models
por: Zhong, Qihuang, et al.
Publicado: (2026)

GenderBias-\emph{VL}: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing
por: Xiao, Yisong, et al.
Publicado: (2024)

Fairness Mediator: Neutralize Stereotype Associations to Mitigate Bias in Large Language Models
por: Xiao, Yisong, et al.
Publicado: (2025)

Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor
por: Wei, Shaokui, et al.
Publicado: (2024)