ACEBench: Who Wins the Match Point in Tool Usage?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Chen, Hao, Xinlong, Liu, Weiwen, Huang, Xu, Zeng, Xingshan, Yu, Shuai, Li, Dexun, Wang, Shuai, Gan, Weinan, Huang, Yuefeng, Liu, Wulong, Wang, Xinzhi, Lian, Defu, Yin, Baoqun, Wang, Yasheng, Liu, Wu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909914850394112
author Chen, Chen
Hao, Xinlong
Liu, Weiwen
Huang, Xu
Zeng, Xingshan
Yu, Shuai
Li, Dexun
Wang, Shuai
Gan, Weinan
Huang, Yuefeng
Liu, Wulong
Wang, Xinzhi
Lian, Defu
Yin, Baoqun
Wang, Yasheng
Liu, Wu
author_facet Chen, Chen
Hao, Xinlong
Liu, Weiwen
Huang, Xu
Zeng, Xingshan
Yu, Shuai
Li, Dexun
Wang, Shuai
Gan, Weinan
Huang, Yuefeng
Liu, Wulong
Wang, Xinzhi
Lian, Defu
Yin, Baoqun
Wang, Yasheng
Liu, Wu
contents Large Language Models (LLMs) have demonstrated significant potential in decision-making and reasoning, particularly when integrated with various tools to effectively solve complex problems. However, existing benchmarks for evaluating LLMs' tool usage face several limitations: (1) limited evaluation scenarios, often lacking assessments in real multi-turn dialogue contexts; (2) narrow evaluation dimensions, with insufficient detailed assessments of how LLMs use tools; and (3) reliance on LLMs or real API executions for evaluation, which introduces significant overhead. To address these challenges, we introduce ACEBench, a comprehensive benchmark for assessing tool usage in LLMs. ACEBench categorizes data into three primary types based on evaluation methodology: Normal, Special, and Agent. "Normal" evaluates tool usage in basic scenarios; "Special" evaluates tool usage in situations with ambiguous or incomplete instructions; "Agent" evaluates tool usage through multi-agent interactions to simulate real-world, multi-turn dialogues. We conducted extensive experiments using ACEBench, analyzing various LLMs in-depth and providing a more granular examination of error causes across different data types.
format Preprint
id arxiv_https___arxiv_org_abs_2501_12851
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ACEBench: Who Wins the Match Point in Tool Usage?
Chen, Chen
Hao, Xinlong
Liu, Weiwen
Huang, Xu
Zeng, Xingshan
Yu, Shuai
Li, Dexun
Wang, Shuai
Gan, Weinan
Huang, Yuefeng
Liu, Wulong
Wang, Xinzhi
Lian, Defu
Yin, Baoqun
Wang, Yasheng
Liu, Wu
Computation and Language
Large Language Models (LLMs) have demonstrated significant potential in decision-making and reasoning, particularly when integrated with various tools to effectively solve complex problems. However, existing benchmarks for evaluating LLMs' tool usage face several limitations: (1) limited evaluation scenarios, often lacking assessments in real multi-turn dialogue contexts; (2) narrow evaluation dimensions, with insufficient detailed assessments of how LLMs use tools; and (3) reliance on LLMs or real API executions for evaluation, which introduces significant overhead. To address these challenges, we introduce ACEBench, a comprehensive benchmark for assessing tool usage in LLMs. ACEBench categorizes data into three primary types based on evaluation methodology: Normal, Special, and Agent. "Normal" evaluates tool usage in basic scenarios; "Special" evaluates tool usage in situations with ambiguous or incomplete instructions; "Agent" evaluates tool usage through multi-agent interactions to simulate real-world, multi-turn dialogues. We conducted extensive experiments using ACEBench, analyzing various LLMs in-depth and providing a more granular examination of error causes across different data types.
title ACEBench: Who Wins the Match Point in Tool Usage?
topic Computation and Language
url https://arxiv.org/abs/2501.12851