ExpertPrompting: Instructing Large Language Models to be Distinguished Experts

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Benfeng, Yang, An, Lin, Junyang, Wang, Quan, Zhou, Chang, Zhang, Yongdong, Mao, Zhendong
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908255908790272
author Xu, Benfeng
Yang, An
Lin, Junyang
Wang, Quan
Zhou, Chang
Zhang, Yongdong
Mao, Zhendong
author_facet Xu, Benfeng
Yang, An
Lin, Junyang
Wang, Quan
Zhou, Chang
Zhang, Yongdong
Mao, Zhendong
contents The answering quality of an aligned large language model (LLM) can be drastically improved if treated with proper crafting of prompts. In this paper, we propose ExpertPrompting to elicit the potential of LLMs to answer as distinguished experts. We first utilize In-Context Learning to automatically synthesize detailed and customized descriptions of the expert identity for each specific instruction, and then ask LLMs to provide answer conditioned on such agent background. Based on this augmented prompting strategy, we produce a new set of instruction-following data using GPT-3.5, and train a competitive open-source chat assistant called ExpertLLaMA. We employ GPT4-based evaluation to show that 1) the expert data is of significantly higher quality than vanilla answers, and 2) ExpertLLaMA outperforms existing open-source opponents and achieves 96\% of the original ChatGPT's capability. All data and the ExpertLLaMA model will be made publicly available at https://github.com/OFA-Sys/ExpertLLaMA.
format Preprint
id arxiv_https___arxiv_org_abs_2305_14688
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle ExpertPrompting: Instructing Large Language Models to be Distinguished Experts
Xu, Benfeng
Yang, An
Lin, Junyang
Wang, Quan
Zhou, Chang
Zhang, Yongdong
Mao, Zhendong
Computation and Language
Artificial Intelligence
The answering quality of an aligned large language model (LLM) can be drastically improved if treated with proper crafting of prompts. In this paper, we propose ExpertPrompting to elicit the potential of LLMs to answer as distinguished experts. We first utilize In-Context Learning to automatically synthesize detailed and customized descriptions of the expert identity for each specific instruction, and then ask LLMs to provide answer conditioned on such agent background. Based on this augmented prompting strategy, we produce a new set of instruction-following data using GPT-3.5, and train a competitive open-source chat assistant called ExpertLLaMA. We employ GPT4-based evaluation to show that 1) the expert data is of significantly higher quality than vanilla answers, and 2) ExpertLLaMA outperforms existing open-source opponents and achieves 96\% of the original ChatGPT's capability. All data and the ExpertLLaMA model will be made publicly available at https://github.com/OFA-Sys/ExpertLLaMA.
title ExpertPrompting: Instructing Large Language Models to be Distinguished Experts
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2305.14688