Security Risk of Misalignment between Text and Image in Multi-modal Model
"攻击方法" 相关文章
Chain-of-Thought Hijacking
A New Type of Adversarial Examples
ADMIT: Few-shot Knowledge Poisoning Attacks on RAG-based Fact Checking
SASER: Stego attacks on open-source LLMs
SECA: Semantically Equivalent and Coherent Attacks for Eliciting LLM Hallucinations
Podcast: Weaponizing Corporate Intel. This Time, It’s Personal!
USENIX Security 25 | TEE+Legacy GPU混合算力下的模型隐私安全攻防
Clone What You Can't Steal: Black-Box LLM Replication via Logit Leakage and Distillation
Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics
人大&港科大揭示大模型重要安全风险漏洞：利用概念激活向量破解大模型的安全对齐｜NeurIPS 2024
A New Google DeepMind Research Reveals a New Kind of Vulnerability that Could Leak User Prompts in MoE Model
「推安早报」1017 | 域安全、红蓝工具节选
