Image de couverture de livre

Harmless Yet Harmful: Neutral Prompting Attacks for Stealthy Hallucination Steering in Agent Skills

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Hsu, Chia-Yi, Yu, Chia-Mu, Huang, Chun-Ying, Sakuma, Jun
Format:	Preprint
Publié:	2026
Sujets:	Cryptography and Security Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Trust Me, Import This: Dependency Steering Attacks via Malicious Agent Skills
par: Liu, Yiyong, et autres
Publié: (2026)

BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors
par: Hsu, Chia-Yi, et autres
Publié: (2025)

Data Poisoning Attacks to Locally Differentially Private Range Query Protocols
par: Liao, Ting-Wei, et autres
Publié: (2025)

Poisoning Attacks to Local Differential Privacy Protocols for Trajectory Data
par: Hsu, I-Jung, et autres
Publié: (2025)

Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on Graph Classification
par: Wang, Xiaobao, et autres
Publié: (2025)

Beyond Full Poisoning: Effective Availability Attacks with Partial Perturbation
par: Zhe, Yu, et autres
Publié: (2024)

Stealthy and Adjustable Text-Guided Backdoor Attacks on Multimodal Pretrained Models
par: Zhang, Yiyang, et autres
Publié: (2026)

Stealthy Adversarial Attacks on Stochastic Multi-Armed Bandits
par: Wang, Zhiwei, et autres
Publié: (2024)

Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
par: Schmotz, David, et autres
Publié: (2026)

SilentStriker:Toward Stealthy Bit-Flip Attacks on Large Language Models
par: Xu, Haotian, et autres
Publié: (2025)

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective
par: Chung, Ming-Yu, et autres
Publié: (2023)

IU: Imperceptible Universal Backdoor Attack
par: Lin, Hsin, et autres
Publié: (2026)

Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models
par: Xu, Yuancheng, et autres
Publié: (2024)

Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning
par: Lyu, Xiaoting, et autres
Publié: (2024)

SDBA: A Stealthy and Long-Lasting Durable Backdoor Attack in Federated Learning
par: Choe, Minyeong, et autres
Publié: (2024)

Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents
par: Zhan, Qiusi, et autres
Publié: (2025)

Krait: A Backdoor Attack Against Graph Prompt Tuning
par: Song, Ying, et autres
Publié: (2024)

The 'Sure' Trap: Multi-Scale Poisoning Analysis of Stealthy Compliance-Only Backdoors in Fine-Tuned Large Language Models
par: Tan, Yuting, et autres
Publié: (2025)

BLAST: A Stealthy Backdoor Leverage Attack against Cooperative Multi-Agent Deep Reinforcement Learning based Systems
par: Fang, Jing, et autres
Publié: (2025)

Stealthy Poisoning Attacks Bypass Defenses in Regression Settings
par: Carnerero-Cano, Javier, et autres
Publié: (2026)

Approaching the Harm of Gradient Attacks While Only Flipping Labels
par: El-Kabid, Abdessamad, et autres
Publié: (2025)

AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
par: Debenedetti, Edoardo, et autres
Publié: (2024)

Data Taggants: Dataset Ownership Verification via Harmless Targeted Data Poisoning
par: Bouaziz, Wassim, et autres
Publié: (2024)

State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
par: Guo, Ji, et autres
Publié: (2026)

A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks
par: Hossain, S M Asif, et autres
Publié: (2025)

Disrupting Model Merging: A Parameter-Level Defense Without Sacrificing Accuracy
par: Junhao, Wei, et autres
Publié: (2025)

PIShield: Detecting Prompt Injection Attacks via Intrinsic LLM Features
par: Zou, Wei, et autres
Publié: (2025)

Graph-Aware Stealthy Poison-Text Backdoors for Text-Attributed Graphs
par: Luo, Qi, et autres
Publié: (2026)

SteganoBackdoor: Stealthy and Data-Efficient Backdoor Attacks on Language Models
par: Xue, Eric, et autres
Publié: (2025)

Attention Tracker: Detecting Prompt Injection Attacks in LLMs
par: Hung, Kuo-Han, et autres
Publié: (2024)

Leveraging Soft Prompts for Privacy Attacks in Federated Prompt Tuning
par: Nguyen, Quan Minh, et autres
Publié: (2026)

AEGIS : Automated Co-Evolutionary Framework for Guarding Prompt Injections Schema
par: Liu, Ting-Chun, et autres
Publié: (2025)

Adversarial Prompt Evaluation: Systematic Benchmarking of Guardrails Against Prompt Input Attacks on LLMs
par: Zizzo, Giulio, et autres
Publié: (2025)

Understanding Data Importance in Machine Learning Attacks: Does Valuable Data Pose Greater Harm?
par: Wen, Rui, et autres
Publié: (2024)

XOXO: Stealthy Cross-Origin Context Poisoning Attacks against AI Coding Assistants
par: Štorek, Adam, et autres
Publié: (2025)

Trident: Improving Malware Detection with LLMs and Behavioral Features
par: Saul, Rebecca, et autres
Publié: (2026)

Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
par: Huang, Tiansheng, et autres
Publié: (2024)

Stealthy Imitation: Reward-guided Environment-free Policy Stealing
par: Zhuang, Zhixiong, et autres
Publié: (2024)

Are You Using Reliable Graph Prompts? Trojan Prompt Attacks on Graph Neural Networks
par: Lin, Minhua, et autres
Publié: (2024)

Neutral Agent-based Adversarial Policy Learning against Deep Reinforcement Learning in Multi-party Open Systems
par: Peng, Qizhou, et autres
Publié: (2025)