一作ACL 2026Poster· Natural Language Processing
C ache P rune: Teaching LLM s What Not to Follow via KV -Cache Editing
Rui Wang · Junda Wu · Yu Xia · Tong Yu · Ruiyi Zhang · Ryan A. Rossi · Subrata Mitra · Lina Yao · Julian J. McAuley
摘要
Large Language Models (LLMs) are susceptible to indirect prompt injection attack, where the model inadvertently responds to instructions injected into the prompt context. This vulnerability stems from LLMs’ inability to distinguish between data and instructions within a prompt. We propose CachePr...