# information-leakage
標記為「information-leakage」的 10 篇文章
Activation Analysis & Hidden State Exploitation
Reading model internals via hidden state extraction, logprob probing, refusal direction analysis, and activation steering techniques.
Lab: GPU Side-Channel Attacks
Demonstrate information leakage through GPU memory residuals and timing side channels, showing how shared GPU infrastructure can expose sensitive model data.
Model Caching Security
Security risks of caching in AI inference — covering KV cache attacks, prompt caching exploitation, shared cache information leakage, cache poisoning, and secure caching configurations.
Cross-Modal Information Leakage
Techniques for extracting sensitive information through modality boundary violations, including system prompt leakage via image descriptions, context window dumping through multi-modal queries, and side-channel attacks across modalities.
Timing Side-Channel Attack Walkthrough
Extract information from LLM applications through timing differences in response generation and safety filter processing.
Activation Analysis & Hidden State 利用ation
Reading model internals via hidden state extraction, logprob probing, refusal direction analysis, and activation steering techniques.
實驗室: GPU Side-Channel 攻擊s
Demonstrate information leakage through GPU memory residuals and timing side channels, showing how shared GPU infrastructure can expose sensitive model data.
模型快取安全
AI 推論中快取的安全風險——涵蓋 KV 快取攻擊、提示詞快取利用、共享快取資訊洩漏、快取投毒與安全快取設定。
Cross-Modal Information Leakage
Techniques for extracting sensitive information through modality boundary violations, including system prompt leakage via image descriptions, context window dumping through multi-modal queries, and side-channel attacks across modalities.
Timing Side-Channel 攻擊 導覽
Extract information from LLM applications through timing differences in response generation and safety filter processing.