Dive into Ambiguity: A*-Inspired Multi-Agents Commonsense Obfuscation Attack on LLM Prompts 文章

ArXiv CS.AI2026-06-02NEWSen作者: Boxuan Wang, Zhuoyun Li, Xiaowei Huang, Yi Dong
查看原文 →
关系图谱

Dive into Ambiguity: A*-Inspired Multi-Agents Commonsense Obfuscation Attack on LLM Prompts · 相关技术

相关技术

reverse simulated annealing scheduleHierarchical Rewrite StrategyA*-inspired Factual Error Induction Frameworklarge language models