Automatic Instruction Evolving for Large Language Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zeng, Weihao, Xu, Can, Zhao, Yingxiu, Lou, Jian-Guang, Chen, Weizhu
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909214855659520
author Zeng, Weihao
Xu, Can
Zhao, Yingxiu
Lou, Jian-Guang
Chen, Weizhu
author_facet Zeng, Weihao
Xu, Can
Zhao, Yingxiu
Lou, Jian-Guang
Chen, Weizhu
contents Fine-tuning large pre-trained language models with Evol-Instruct has achieved encouraging results across a wide range of tasks. However, designing effective evolving methods for instruction evolution requires substantial human expertise. This paper proposes Auto Evol-Instruct, an end-to-end framework that evolves instruction datasets using large language models without any human effort. The framework automatically analyzes and summarizes suitable evolutionary strategies for the given instruction data and iteratively improves the evolving method based on issues exposed during the instruction evolution process. Our extensive experiments demonstrate that the best method optimized by Auto Evol-Instruct outperforms human-designed methods on various benchmarks, including MT-Bench, AlpacaEval, GSM8K, and HumanEval.
format Preprint
id arxiv_https___arxiv_org_abs_2406_00770
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Automatic Instruction Evolving for Large Language Models
Zeng, Weihao
Xu, Can
Zhao, Yingxiu
Lou, Jian-Guang
Chen, Weizhu
Computation and Language
Artificial Intelligence
Fine-tuning large pre-trained language models with Evol-Instruct has achieved encouraging results across a wide range of tasks. However, designing effective evolving methods for instruction evolution requires substantial human expertise. This paper proposes Auto Evol-Instruct, an end-to-end framework that evolves instruction datasets using large language models without any human effort. The framework automatically analyzes and summarizes suitable evolutionary strategies for the given instruction data and iteratively improves the evolving method based on issues exposed during the instruction evolution process. Our extensive experiments demonstrate that the best method optimized by Auto Evol-Instruct outperforms human-designed methods on various benchmarks, including MT-Bench, AlpacaEval, GSM8K, and HumanEval.
title Automatic Instruction Evolving for Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.00770