Shared Lexical Task Representations Explain Behavioral Variability In LLMs

ArXi:2604.22027v1 Announce Type: cross One of the most common complaints about large language models (LLMs) is their prompt sensitivity -- that is, the fact that their ability to perform a task or provide a correct answer to a question can depend unpredictably on the way the question is posed. We investigate this variation by comparing two very different but commonly-used styles of prompting: instruction-based prompts, which describe the task in natural language, and example-based prompts, which provide in-context few-shot nstration pairs to illustrate the task.