BioTool: A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language Models

2026-05-07

AI Summary

This study addresses the challenge of inaccurate tool invocation by large language models (LLMs) in specialized domains such as biomedicine, which hinders their practical deployment. To bridge this gap, the authors introduce the first large-scale, human-validated dataset for biomedical tool calling, encompassing 34 tools from major platforms including NCBI, Ensembl, and UniProt, and spanning key areas such as variant analysis, genomics, and proteomics. Leveraging 7,040 expert-annotated query–API call pairs, they perform supervised fine-tuning on a 4-billion-parameter LLM. Experimental results demonstrate that the fine-tuned model significantly outperforms state-of-the-art commercial models, including GPT-5.1, in tool-calling accuracy. Expert evaluations further confirm that this improvement translates into enhanced quality in downstream question-answering tasks.

Abstract

Despite the success of large language models (LLMs) on general-purpose tasks, their performance in highly specialized domains such as biomedicine remains unsatisfactory. A key limitation is the inability of LLMs to effectively leverage biomedical tools, which clinical experts and biomedical researchers rely on extensively in daily workflows. While recent general-domain tool-calling datasets have substantially improved the capabilities of LLM agents, existing efforts in the biomedical domain largely rely on in-context learning and restrict models to a small set of tools. To address this gap, we introduce BioTool, a comprehensive biomedical tool-calling dataset designed for fine-tuning LLMs. BioTool comprises 34 frequently used tools collected from the NCBI, Ensembl, and UniProt databases, along with 7,040 high-quality, human-verified query-API call pairs spanning variation, genomics, proteomics, evolution, and general biology. Fine-tuning a 4-billion-parameter LLM on BioTool yields substantial improvements in biomedical tool-calling performance, outperforming cutting-edge commercial LLMs such as GPT-5.1. Furthermore, human expert evaluations demonstrate that integrating a BioTool-fine-tuned tool caller significantly improves downstream answer quality compared to the same LLM without tool usage, highlighting the effectiveness of BioTool in enhancing the biomedical capabilities of LLMs. The full dataset and evaluation code are available at https://github.com/gxx27/BioTool

Problem

• large language models
• biomedical tools
• tool calling
• specialized domains
• biomedicine

Innovation

• tool calling
• biomedical LLMs
• fine-tuning dataset
• API integration
• domain-specific reasoning