@inproceedings{zhao-etal-2025-multifaceted,
    title = "Multifaceted Evaluation of Audio-Visual Capability for {MLLM}s: Effectiveness, Efficiency, Generalizability and Robustness",
    author = "Zhao, Yusheng  and
      Luo, Xiao  and
      Luo, Junyu  and
      Zhang, Weizhi  and
      Xiao, Zhiping  and
      Ju, Wei  and
      Yu, Philip S.  and
      Zhang, Ming",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.54/",
    doi = "10.18653/v1/2025.findings-emnlp.54",
    pages = "1026--1041",
    ISBN = "979-8-89176-335-7",
    abstract = "Multi-modal large language models (MLLMs) have recently achieved great success in processing and understanding information from diverse modalities (e.g., text, audio, and visual signals). Despite their growing popularity, there remains a lack of comprehensive evaluation measuring the audio-visual capabilities of these models, especially in diverse scenarios (e.g., distribution shifts and adversarial attacks). In this paper, we present a multifaceted evaluation of the audio-visual capability of MLLMs, focusing on four key dimensions: effectiveness, efficiency, generalizability, and robustness. Through extensive experiments, we find that MLLMs exhibit strong zero-shot and few-shot generalization abilities, enabling them to achieve great performance with limited data. However, their success relies heavily on the vision modality, which impairs performance when visual input is corrupted or missing. Additionally, while MLLMs are susceptible to adversarial samples, they demonstrate greater robustness compared to traditional models. The experimental results and our observations provide new insights into the audio-visual capabilities of MLLMs, highlighting areas for improvement and offering guidance for future research."
}