ToolDreamer: Instilling LLM Reasoning Into Tool Retrievers

TL;DR

ToolDreamer enhances tool retrieval for LLMs by generating synthetic descriptions to better match user queries, enabling more effective tool selection without exceeding context limits.

Contribution

It introduces a novel framework that conditions retrievers on synthetic tool descriptions generated by LLMs, improving retrieval performance for large tool sets.

Findings

Improves sparse and dense retriever performance

Enhances tool retrieval accuracy with synthetic descriptions

Demonstrates flexibility across different retriever models

Abstract

Tool calling has become increasingly popular for Large Language Models (LLMs). However, for large tool sets, the resulting tokens would exceed the LLM's context window limit, making it impossible to include every tool. Hence, an external retriever is used to provide LLMs with the most relevant tools for a query. Existing retrieval models rank tools based on the similarity between a user query and a tool description (TD). This leads to suboptimal retrieval as user requests are often poorly aligned with the language of TD. To remedy the issue, we propose ToolDreamer, a framework to condition retriever models to fetch tools based on hypothetical (synthetic) TD generated using an LLM, i.e., description of tools that the LLM feels will be potentially useful for the query. The framework enables a more natural alignment between queries and tools within the language space of TD's. We apply ToolDreamer on the ToolRet dataset and show that our method improves the performance of sparse and dense retrievers with and without training, thus showcasing its flexibility. Through our proposed framework, our aim is to offload a portion of the reasoning burden to the retriever so that the LLM may effectively handle a large collection of tools without inundating its context window.