Cordyceps: Covert Control Attacks on LLMs via Data Poisoning 文章

ArXiv CS.AI2026-05-27NEWSen作者: Zedian Shao, Charles Fleming, Teodora Baluta
大语言模型
查看原文 →
关系图谱

Cordyceps: Covert Control Attacks on LLMs via Data Poisoning · 相关技术

相关技术

LLMORMODElanguage modelrsaprompt injectionevalUCTSynthetic Aperture RadarStraight-Through EstimatorSEMReferring expression comprehension (REC)PromptParts-of-Speech (POS) tagsOWLNLIMaliLMMlarge language modelsHISForEffort Metric AttentionDecodeCamouflaged object detectionCISARGANN