MTU-Bench: A Multi-granularity Tool-Use Benchmark for Large Language Models
Alibaba Group · Beijing University of Posts and Telecommunications · Beijing University of Aeronautics and Astronautics · Nanjing University · University of the Chinese Academy of Sciences · Kuaishou Technology · Institute of automation, Chinese academy of science · University of Waterloo · Institute of automation, Chinese academy of science, Chinese Academy of Sciences · ALIBABA
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Large Language Models (LLMs) have displayed massive improvements in reason- ing and decision-making skills and can hold natural conversations with users. Recently, many tool-use benchmark datasets have been proposed. However, existing datasets have the following limitations: (1). Insufficient evaluation scenarios (e.g., only cover limited tool-use scenes). (2). Extensive evaluation costs (e.g., GPT API costs). To address these limitations, in this work, we propose a multi-granularity tool-use benchmark for large language models called MTU-Bench. For the "multi-granularity" property, our MTU-Bench covers five tool usage scenes (i.e., single-turn and single-tool, single-turn and multiple-tool, multiple-turn and single-tool, multiple-turn and multiple-tool, and out-of-distribution tasks). Besides, all evaluation metrics of our MTU-Bench are based on the prediction results and the ground truth without using any GPT or human evaluation metrics. Moreover, our MTU-Bench is collected by transforming existing high-quality datasets to simulate real-world tool usage scenarios, and we also propose an instruction dataset called MTU-Instruct data to enhance the tool-use abilities of existing LLMs. Comprehensive experimental results demonstrate the effectiveness of our MTU-Bench.