Image de couverture de livre

Transferable Direct Prompt Injection via Activation-Guided MCMC Sampling

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Minghui, Zhang, Hao, Zhang, Yechao, Wan, Wei, Hu, Shengshan, Xiaobing, pei, Wang, Jing
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Transferable Adversarial Facial Images for Privacy Protection
par: Li, Minghui, et autres
Publié: (2024)

HiF-DTA: Hierarchical Feature Learning Network for Drug-Target Affinity Prediction
par: Li, Minghui, et autres
Publié: (2025)

ViDTA: Enhanced Drug-Target Affinity Prediction via Virtual Graph Nodes and Attention-based Feature Fusion
par: Li, Minghui, et autres
Publié: (2024)

MARS: A Malignity-Aware Backdoor Defense in Federated Learning
par: Wan, Wei, et autres
Publié: (2025)

PB-UAP: Hybrid Universal Adversarial Attack For Image Segmentation
par: Song, Yufei, et autres
Publié: (2024)

Spa-VLM: Stealthy Poisoning Attacks on RAG-based VLM
par: Yu, Lei, et autres
Publié: (2025)

DarkSAM: Fooling Segment Anything Model to Segment Nothing
par: Zhou, Ziqi, et autres
Publié: (2024)

JANUS: A Dual-Constraint Generative Framework for Stealthy Node Injection Attacks
par: Zhang, Jiahao, et autres
Publié: (2025)

Detecting and Corrupting Convolution-based Unlearnable Examples
par: Li, Minghui, et autres
Publié: (2023)

Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling
par: Zheng, Guangmin, et autres
Publié: (2024)

Why Does Little Robustness Help? A Further Step Towards Understanding Adversarial Transferability
par: Zhang, Yechao, et autres
Publié: (2023)

Learning to Inject: Automated Prompt Injection via Reinforcement Learning
par: Chen, Xin, et autres
Publié: (2026)

Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection
par: Chen, Meng, et autres
Publié: (2026)

Secure Transfer Learning: Training Clean Models Against Backdoor in (Both) Pre-trained Encoders and Downstream Datasets
par: Zhang, Yechao, et autres
Publié: (2025)

Exploring Gradient-Guided Masked Language Model to Detect Textual Adversarial Attacks
par: Zhang, Xiaomei, et autres
Publié: (2025)

EVA: Red-Teaming GUI Agents via Evolving Indirect Prompt Injection
par: Lu, Yijie, et autres
Publié: (2025)

AdapTools: Adaptive Tool-based Indirect Prompt Injection Attacks on Agentic LLMs
par: Wang, Che, et autres
Publié: (2026)

ICON: Indirect Prompt Injection Defense for Agents based on Inference-Time Correction
par: Wang, Che, et autres
Publié: (2026)

To Protect the LLM Agent Against the Prompt Injection Attack with Polymorphic Prompt
par: Wang, Zhilong, et autres
Publié: (2025)

Prompt to Transfer: Sim-to-Real Transfer for Traffic Signal Control with Prompt Learning
par: Da, Longchao, et autres
Publié: (2023)

Automatic and Universal Prompt Injection Attacks against Large Language Models
par: Liu, Xiaogeng, et autres
Publié: (2024)

ConsPrompt: Exploiting Contrastive Samples for Fewshot Prompt Learning
par: Weng, Jinta, et autres
Publié: (2022)

Metric-Fair Prompting: Treating Similar Samples Similarly
par: Wang, Jing, et autres
Publié: (2025)

Give Them an Inch and They Will Take a Mile:Understanding and Measuring Caller Identity Confusion in MCP-Based AI Systems
par: Huang, Yuhang, et autres
Publié: (2026)

NLP Sampling: Combining MCMC and NLP Methods for Diverse Constrained Sampling
par: Toussaint, Marc, et autres
Publié: (2024)

ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack
par: Li, Hao, et autres
Publié: (2026)

PromptBridge: Cross-Model Prompt Transfer for Large Language Models
par: Wang, Yaxuan, et autres
Publié: (2025)

QueryIPI: Query-agnostic Indirect Prompt Injection on Coding Agents
par: Xie, Yuchong, et autres
Publié: (2025)

Robot Collapse: Supply Chain Backdoor Attacks Against VLM-based Robotic Manipulation
par: Wang, Xianlong, et autres
Publié: (2024)

ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injection
par: Zhao, Wei, et autres
Publié: (2026)

Small Shifts, Large Gains: Unlocking Traditional TSP Heuristic Guided-Sampling via Unsupervised Neural Instance Modification
par: Huang, Wei, et autres
Publié: (2026)

HiStyle: Hierarchical Style Embedding Predictor for Text-Prompt-Guided Controllable Speech Synthesis
par: Zhang, Ziyu, et autres
Publié: (2025)

Physical Prompt Injection Attacks on Large Vision-Language Models
par: Ling, Chen, et autres
Publié: (2026)

InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models
par: Li, Hao, et autres
Publié: (2024)

Uncertainty-Aware Metabolic Stability Prediction with Dual-View Contrastive Learning
par: Guo, Peijin, et autres
Publié: (2025)

Prompt2Auto: From Motion Prompt to Automated Control via Geometry-Invariant One-Shot Gaussian Process Learning
par: Yang, Zewen, et autres
Publié: (2025)

PromptArmor: Simple yet Effective Prompt Injection Defenses
par: Shi, Tianneng, et autres
Publié: (2025)

SecInfer: Preventing Prompt Injection via Inference-time Scaling
par: Liu, Yupei, et autres
Publié: (2025)

Bypassing Prompt Injection Detectors through Evasive Injections
par: Rahman, Md Jahedur, et autres
Publié: (2026)

PromptLocate: Localizing Prompt Injection Attacks
par: Jia, Yuqi, et autres
Publié: (2025)