Saved in:
Bibliographic Details
Main Authors: Xu, Qiancheng, Li, Yongqi, Xia, Heming, Li, Wenjie
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.17465
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929519120613376
author Xu, Qiancheng
Li, Yongqi
Xia, Heming
Li, Wenjie
author_facet Xu, Qiancheng
Li, Yongqi
Xia, Heming
Li, Wenjie
contents Tool learning aims to enhance and expand large language models' (LLMs) capabilities with external tools, which has gained significant attention recently. Current methods have shown that LLMs can effectively handle a certain amount of tools through in-context learning or fine-tuning. However, in real-world scenarios, the number of tools is typically extensive and irregularly updated, emphasizing the necessity for a dedicated tool retrieval component. Tool retrieval is nontrivial due to the following challenges: 1) complex user instructions and tool descriptions; 2) misalignment between tool retrieval and tool usage models. To address the above issues, we propose to enhance tool retrieval with iterative feedback from the large language model. Specifically, we prompt the tool usage model, i.e., the LLM, to provide feedback for the tool retriever model in multi-round, which could progressively improve the tool retriever's understanding of instructions and tools and reduce the gap between the two standalone components. We build a unified and comprehensive benchmark to evaluate tool retrieval models. The extensive experiments indicate that our proposed approach achieves advanced performance in both in-domain evaluation and out-of-domain evaluation.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17465
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Tool Retrieval with Iterative Feedback from Large Language Models
Xu, Qiancheng
Li, Yongqi
Xia, Heming
Li, Wenjie
Computation and Language
Artificial Intelligence
Tool learning aims to enhance and expand large language models' (LLMs) capabilities with external tools, which has gained significant attention recently. Current methods have shown that LLMs can effectively handle a certain amount of tools through in-context learning or fine-tuning. However, in real-world scenarios, the number of tools is typically extensive and irregularly updated, emphasizing the necessity for a dedicated tool retrieval component. Tool retrieval is nontrivial due to the following challenges: 1) complex user instructions and tool descriptions; 2) misalignment between tool retrieval and tool usage models. To address the above issues, we propose to enhance tool retrieval with iterative feedback from the large language model. Specifically, we prompt the tool usage model, i.e., the LLM, to provide feedback for the tool retriever model in multi-round, which could progressively improve the tool retriever's understanding of instructions and tools and reduce the gap between the two standalone components. We build a unified and comprehensive benchmark to evaluate tool retrieval models. The extensive experiments indicate that our proposed approach achieves advanced performance in both in-domain evaluation and out-of-domain evaluation.
title Enhancing Tool Retrieval with Iterative Feedback from Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.17465