NAACL 2025findings2 citations

MMAU: A Holistic Benchmark of Agent Capabilities Across Diverse Domains

Guoli Yin, Haoping Bai, Shuang Ma, Feng Nan, Yanchao Sun, Zhaoyang Xu, Shen Ma, Jiarui Lu

Abstract

Recent advances in large language models (LLMs) have increased the demand for comprehensive benchmarks to evaluate their capabilities as human-like agents. Existing benchmarks, while useful, often focus on specific application scenarios, emphasizing task completion but failing to dissect the underlying skills that drive these outcomes. This lack of granularity makes it difficult to deeply discern where failures stem from. Additionally, setting up these environments requires considerable effort, and issues of unreliability and reproducibility sometimes arise, especially in interactive tasks. To address these limitations, we introduce the Massive Multitask Agent Understanding (MMAU) benchmark, featuring comprehensive offline tasks that eliminate the need for complex environment setups. It evaluate models across five domains, including Tool-use, Directed Acyclic Graph (DAG) QA, Data Science and Machine Learning coding, Contest-level programming and Mathematics, and covering five essential capabilities: Understanding, Reasoning, Planning, Problem-solving, and Self-correction. With a total of 20 meticulously designed tasks encompassing over 3K distinct prompts, MMAU provides a comprehensive framework for evaluating the strengths and limitations of LLM agents. By testing 20 representative models on MMAU, we provide deep and insightful analyses. Ultimately, MMAU not only sheds light on the capabilities and limitations of LLM agents but also enhances the interpretability of their performance.

BibTeX
@inproceedings{yin-etal-2025-mmau,
    title = "{MMAU}: A Holistic Benchmark of Agent Capabilities Across Diverse Domains",
    author = "Yin, Guoli  and
      Bai, Haoping  and
      Ma, Shuang  and
      Nan, Feng  and
      Sun, Yanchao  and
      Xu, Zhaoyang  and
      Ma, Shen  and
      Lu, Jiarui  and
      Kong, Xiang  and
      Zhang, Aonan  and
      Yap, Dian Ang  and
      Zhang, Yizhe  and
      Ahnert, Karsten  and
      Kamath, Vik  and
      Berglund, Mathias  and
      Walsh, Dominic  and
      Gindele, Tobias  and
      Wiest, Juergen  and
      Lai, Zhengfeng  and
      Wang, Xiaoming Simon  and
      Shan, Jiulong  and
      Cao, Meng  and
      Pang, Ruoming  and
      Wang, Zirui",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.findings-naacl.267/",
    pages = "4737--4765",
    ISBN = "979-8-89176-195-7"
}