一作ACL 2026Findings· Findings of ACL
Beyond Prompt Engineering: A Systematic Analysis of Prompt Lexical Sensitivity and Its Impacts on Quality
Qipeng Xie · Zi Liang · Jiafei Wu · Yufei Chen · Weizheng Wang · Wenao Ma · Zhong Ming · Haiqin Yang · Kaishun Wu
摘要
Large Language Models (LLMs) often exhibit extreme sensitivity to surface-level prompt variations, where minor lexical perturbations trigger disproportionate performance fluctuations. Moving beyond black-box optimization or coarse-grained templates, we conduct the first analysis of n-gram token-l...