ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Use

TL;DR

ToolHop is a new benchmark dataset designed to evaluate large language models' multi-hop tool use capabilities, revealing significant challenges and variability among models, with GPT-4o achieving around 49% accuracy.

Contribution

We introduce ToolHop, a comprehensive, query-driven dataset for rigorous evaluation of multi-hop tool use in large language models, including diverse queries, meaningful dependencies, and verifiable answers.

Findings

GPT-4o achieves 49.04% accuracy on ToolHop.

Significant challenges remain for LLMs in multi-hop tool use.

Variations in tool-use strategies across model families were observed.

Abstract

Effective evaluation of multi-hop tool use is critical for analyzing the understanding, reasoning, and function-calling capabilities of large language models (LLMs). However, progress has been hindered by a lack of reliable evaluation datasets. To address this, we present ToolHop, a dataset comprising 995 user queries and 3,912 associated tools, specifically designed for rigorous evaluation of multi-hop tool use. ToolHop ensures diverse queries, meaningful interdependencies, locally executable tools, detailed feedback, and verifiable answers through a novel query-driven data construction approach that includes tool creation, document refinement, and code generation. We evaluate 14 LLMs across five model families (i.e., LLaMA3.1, Qwen2.5, Gemini1.5, Claude3.5, and GPT), uncovering significant challenges in handling multi-hop tool-use scenarios. The leading model, GPT-4o, achieves an accuracy of 49.04%, underscoring substantial room for improvement. Further analysis reveals variations in tool-use strategies for various families, offering actionable insights to guide the development of more effective approaches. Code and data can be found in https://huggingface.co/datasets/bytedance-research/ToolHop.