Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges
This addresses a gap in evaluating language agents for real-world multi-turn applications, though it is incremental as it builds on existing tool-use benchmarks.
The authors tackled the lack of benchmarks for stateful tool use in multi-turn dialogues by introducing DialogTool, a dataset with six tasks across three stages, and VirtualMobile, an evaluation environment, finding that current state-of-the-art LLMs perform poorly in long-horizon tool interactions.
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.