LangGPT: Rethinking Structured Reusable Prompt Design Framework for LLMs from the Programming Language

arXiv:2402.16929v1 Announce Type: cross
Abstract: LLMs have demonstrated commendable performance across diverse domains. Nevertheless, formulating high-quality prompts to effectively instruct LLMs poses a challenge for non-AI experts. Existing research in prompt engineering suggests somewhat fragmented optimization principles and designs empirically dependent prompt optimizers. Unfortunately, these endeavors lack a structured design template, incurring high learning costs and resulting in low reusability. Inspired by structured reusable programming languages, we propose LangGPT, a dual-layer prompt design framework as the programming language for LLMs. …

abstract ai experts arxiv challenge cs.ai cs.cl cs.pl cs.se design designs diverse domains engineering experts framework language llms optimization performance programming programming language prompt prompts quality research type