Retrieval Models Aren't Tool-Savvy: Benchmarking Tool Retrieval for Large Language Models
This addresses the bottleneck of tool retrieval in real-world LLM applications, which is incremental but important for enhancing tool-use efficiency.
The paper tackles the problem of tool retrieval for large language models (LLMs) by introducing ToolRet, a benchmark with 7.6k retrieval tasks and 43k tools, and finds that even strong IR models perform poorly, degrading LLM task pass rates; they also provide a 200k-instance training dataset that improves retrieval ability.
Tool learning aims to augment large language models (LLMs) with diverse tools, enabling them to act as agents for solving practical tasks. Due to the limited context length of tool-using LLMs, adopting information retrieval (IR) models to select useful tools from large toolsets is a critical initial step. However, the performance of IR models in tool retrieval tasks remains underexplored and unclear. Most tool-use benchmarks simplify this step by manually pre-annotating a small set of relevant tools for each task, which is far from the real-world scenarios. In this paper, we propose ToolRet, a heterogeneous tool retrieval benchmark comprising 7.6k diverse retrieval tasks, and a corpus of 43k tools, collected from existing datasets. We benchmark six types of models on ToolRet. Surprisingly, even the models with strong performance in conventional IR benchmarks, exhibit poor performance on ToolRet. This low retrieval quality degrades the task pass rate of tool-use LLMs. As a further step, we contribute a large-scale training dataset with over 200k instances, which substantially optimizes the tool retrieval ability of IR models.