Yichen Tang, Weihang Su, Yiqun Liu, Qingyao Ai
The paper introduces a Multi-Field Tool Retrieval framework to improve how Large Language Models select external tools by addressing challenges in tool documentation and user query alignment.
Large Language Models (LLMs) are powerful AI systems that can perform complex tasks by using external tools. However, finding the right tool among many available options can be challenging due to incomplete or inconsistent documentation and differences between user questions and technical tool descriptions. This paper presents a new method called Multi-Field Tool Retrieval, which improves how these models choose tools by considering various aspects like functionality and input/output requirements. The proposed approach has been tested and shown to outperform existing methods, making it more reliable and effective in different situations.
