Saved in:
Bibliographic Details
Main Authors: Wang, Ming, Wu, Wenfang, Gao, Chongyun, Wang, Daling, Feng, Shi, Zhang, Yifei
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2307.15997
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929584205725696
author Wang, Ming
Wu, Wenfang
Gao, Chongyun
Wang, Daling
Feng, Shi
Zhang, Yifei
author_facet Wang, Ming
Wu, Wenfang
Gao, Chongyun
Wang, Daling
Feng, Shi
Zhang, Yifei
contents Large language models (LLMs) have received increasing attention. However, due to the complexity of its capabilities, how to rationally evaluate the capabilities of LLMs is still a task to be solved. We propose the RoCar method, which utilizes the defined basic schemas to randomly construct a task graph and generates natural language evaluation tasks based on the task graph to evaluate the reasoning and memory abilities of LLMs respectively. Due to the very large randomness of the task construction process, it is possible to ensure that none of the LLMs to be tested has directly learned the evaluation tasks, guaranteeing the fairness of the evaluation method.
format Preprint
id arxiv_https___arxiv_org_abs_2307_15997
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle RoCar: A Relationship Network-based Evaluation Method for Large Language Models
Wang, Ming
Wu, Wenfang
Gao, Chongyun
Wang, Daling
Feng, Shi
Zhang, Yifei
Computation and Language
Artificial Intelligence
Large language models (LLMs) have received increasing attention. However, due to the complexity of its capabilities, how to rationally evaluate the capabilities of LLMs is still a task to be solved. We propose the RoCar method, which utilizes the defined basic schemas to randomly construct a task graph and generates natural language evaluation tasks based on the task graph to evaluate the reasoning and memory abilities of LLMs respectively. Due to the very large randomness of the task construction process, it is possible to ensure that none of the LLMs to be tested has directly learned the evaluation tasks, guaranteeing the fairness of the evaluation method.
title RoCar: A Relationship Network-based Evaluation Method for Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2307.15997