Large language models (LLMs) demonstrate exceptional instruct-following ability to complete various downstream tasks. Although this impressive ability makes LLMs flexible task solvers, their performance in solving tasks also heavily relies on instructions. In this paper, we reveal that LLMs are over-sensitive to lexical variations in task instructions, even when the variations are imperceptible to humans. By providing models with neighborhood instructions, which are closely situated in the latent representation space and differ by only one semantically similar word, the performance on downstream tasks can be vastly different. Following this property, we propose a black-box Combinatorial Optimization framework for Prompt Lexical Enhancement (COPLE). COPLE performs iterative lexical optimization according to the feedback from a batch of proxy tasks, using a search strategy related to word influence. Experiments show that even widely-used human-crafted prompts for current benchmarks suffer from the lexical sensitivity of models, and COPLE recovers the declined model ability in both instruct-following and solving downstream tasks.

结合最近的研究，本文发现大型语言模型对任务指令中的词语变化非常敏感，即使对人类来说变化几乎无法察觉。通过在潜在表示空间中提供相近但仅有一个语义上相似的词语的邻居指令，我们发现模型在完成下游任务的性能会有巨大差异。基于这一性质，我们提出了一个黑盒优化框架，即Prompt Lexical Enhancement的组合优化（COPLE）。COPLE根据一批代理任务的反馈进行迭代的词语优化，并使用与词语影响力相关的搜索策略。实验证明，即使是目前流行的人工设计的提示任务，也因模型对词语敏感性而受到影响，而COPLE在遵循指令、解决下游任务方面恢复了模型的性能。

透视LLMs的词汇敏感性：为Prompt增强进行组合优化