Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Hongru, Huang, Wenyu, Wang, Yufei, Xi, Yuanhao, Lu, Jianqiao, Zhang, Huan, Hu, Nan, Liu, Zeming, Pan, Jeff Z., Wong, Kam-Fai
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909615987359744
author Wang, Hongru
Huang, Wenyu
Wang, Yufei
Xi, Yuanhao
Lu, Jianqiao
Zhang, Huan
Hu, Nan
Liu, Zeming
Pan, Jeff Z.
Wong, Kam-Fai
author_facet Wang, Hongru
Huang, Wenyu
Wang, Yufei
Xi, Yuanhao
Lu, Jianqiao
Zhang, Huan
Hu, Nan
Liu, Zeming
Pan, Jeff Z.
Wong, Kam-Fai
contents Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-turn dialogue dataset with stateful tool interactions considering the whole life cycle of tool use, across six key tasks in three stages: 1) \textit{tool creation}; 2) \textit{tool utilization}: tool awareness, tool selection, tool execution; and 3) \textit{role-consistent response}: response generation and role play. Furthermore, we build \texttt{VirtualMobile} -- an embodied virtual mobile evaluation environment to simulate API calls and assess the robustness of the created APIs\footnote{We will use tools and APIs alternatively, there are no significant differences between them in this paper.}. Taking advantage of these artifacts, we conduct comprehensive evaluation on 13 distinct open- and closed-source LLMs and provide detailed analysis at each stage, revealing that the existing state-of-the-art LLMs still cannot perform well to use tools over long horizons.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13328
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges
Wang, Hongru
Huang, Wenyu
Wang, Yufei
Xi, Yuanhao
Lu, Jianqiao
Zhang, Huan
Hu, Nan
Liu, Zeming
Pan, Jeff Z.
Wong, Kam-Fai
Computation and Language
Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-turn dialogue dataset with stateful tool interactions considering the whole life cycle of tool use, across six key tasks in three stages: 1) \textit{tool creation}; 2) \textit{tool utilization}: tool awareness, tool selection, tool execution; and 3) \textit{role-consistent response}: response generation and role play. Furthermore, we build \texttt{VirtualMobile} -- an embodied virtual mobile evaluation environment to simulate API calls and assess the robustness of the created APIs\footnote{We will use tools and APIs alternatively, there are no significant differences between them in this paper.}. Taking advantage of these artifacts, we conduct comprehensive evaluation on 13 distinct open- and closed-source LLMs and provide detailed analysis at each stage, revealing that the existing state-of-the-art LLMs still cannot perform well to use tools over long horizons.
title Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges
topic Computation and Language
url https://arxiv.org/abs/2505.13328