Seal-Tools: Self-Instruct Tool Learning Dataset for Agent Tuning and Detailed Benchmark
This provides a new benchmark for researchers and developers to assess and improve tool learning in AI agents, though it is incremental as it builds on existing tool-learning concepts.
The paper tackles the problem of evaluating tool-calling abilities in large language models by introducing Seal-Tools, a new dataset with self-instruct API-like tools and instances, including hard cases with multiple or nested tool calls, and results show current systems are far from perfect.
This paper presents a new tool learning dataset Seal-Tools, which contains self-instruct API-like tools. Seal-Tools not only offers a large number of tools, but also includes instances which demonstrate the practical application of tools. Seeking to generate data on a large scale while ensuring reliability, we propose a self-instruct method to generate tools and instances, allowing precise control over the process. Moreover, our Seal-Tools contains hard instances that call multiple tools to complete the job, among which some are nested tool callings. For precise and comprehensive evaluation, we use strict format control and design three metrics from different dimensions. Therefore, Seal-Tools can serve as a new benchmark to evaluate the tool-calling ability of LLMs. Finally, we evaluate several prevalent LLMs and our finetuned model on Seal-Tools. The results show that current systems are far from perfect. The code, data and experiment results are available at https://github.com/fairyshine/Seal-Tools .