PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yidong, Yu, Zhuohao, Zeng, Zhengran, Yang, Linyi, Wang, Cunxiang, Chen, Hao, Jiang, Chaoya, Xie, Rui, Wang, Jindong, Xie, Xing, Ye, Wei, Zhang, Shikun, Zhang, Yue
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916258214051840
author Wang, Yidong
Yu, Zhuohao
Zeng, Zhengran
Yang, Linyi
Wang, Cunxiang
Chen, Hao
Jiang, Chaoya
Xie, Rui
Wang, Jindong
Xie, Xing
Ye, Wei
Zhang, Shikun
Zhang, Yue
author_facet Wang, Yidong
Yu, Zhuohao
Zeng, Zhengran
Yang, Linyi
Wang, Cunxiang
Chen, Hao
Jiang, Chaoya
Xie, Rui
Wang, Jindong
Xie, Xing
Ye, Wei
Zhang, Shikun
Zhang, Yue
contents Instruction tuning large language models (LLMs) remains a challenging task, owing to the complexity of hyperparameter selection and the difficulty involved in evaluating the tuned models. To determine the optimal hyperparameters, an automatic, robust, and reliable evaluation benchmark is essential. However, establishing such a benchmark is not a trivial task due to the challenges associated with evaluation accuracy and privacy protection. In response to these challenges, we introduce a judge large language model, named PandaLM, which is trained to distinguish the superior model given several LLMs. PandaLM's focus extends beyond just the objective correctness of responses, which is the main focus of traditional evaluation datasets. It addresses vital subjective factors such as relative conciseness, clarity, adherence to instructions, comprehensiveness, and formality. To ensure the reliability of PandaLM, we collect a diverse human-annotated test dataset, where all contexts are generated by humans and labels are aligned with human preferences. Our results indicate that PandaLM-7B achieves 93.75% of GPT-3.5's evaluation ability and 88.28% of GPT-4's in terms of F1-score on our test dataset. PandaLM enables the evaluation of LLM to be fairer but with less cost, evidenced by significant improvements achieved by models tuned through PandaLM compared to their counterparts trained with default Alpaca's hyperparameters. In addition, PandaLM does not depend on API-based evaluations, thus avoiding potential data leakage. All resources of PandaLM are released at https://github.com/WeOpenML/PandaLM.
format Preprint
id arxiv_https___arxiv_org_abs_2306_05087
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization
Wang, Yidong
Yu, Zhuohao
Zeng, Zhengran
Yang, Linyi
Wang, Cunxiang
Chen, Hao
Jiang, Chaoya
Xie, Rui
Wang, Jindong
Xie, Xing
Ye, Wei
Zhang, Shikun
Zhang, Yue
Computation and Language
Artificial Intelligence
Instruction tuning large language models (LLMs) remains a challenging task, owing to the complexity of hyperparameter selection and the difficulty involved in evaluating the tuned models. To determine the optimal hyperparameters, an automatic, robust, and reliable evaluation benchmark is essential. However, establishing such a benchmark is not a trivial task due to the challenges associated with evaluation accuracy and privacy protection. In response to these challenges, we introduce a judge large language model, named PandaLM, which is trained to distinguish the superior model given several LLMs. PandaLM's focus extends beyond just the objective correctness of responses, which is the main focus of traditional evaluation datasets. It addresses vital subjective factors such as relative conciseness, clarity, adherence to instructions, comprehensiveness, and formality. To ensure the reliability of PandaLM, we collect a diverse human-annotated test dataset, where all contexts are generated by humans and labels are aligned with human preferences. Our results indicate that PandaLM-7B achieves 93.75% of GPT-3.5's evaluation ability and 88.28% of GPT-4's in terms of F1-score on our test dataset. PandaLM enables the evaluation of LLM to be fairer but with less cost, evidenced by significant improvements achieved by models tuned through PandaLM compared to their counterparts trained with default Alpaca's hyperparameters. In addition, PandaLM does not depend on API-based evaluations, thus avoiding potential data leakage. All resources of PandaLM are released at https://github.com/WeOpenML/PandaLM.
title PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2306.05087