@inproceedings{chen-etal-2025-adammeme,
    title = "{A}dam{M}eme: Adaptively Probe the Reasoning Capacity of Multimodal Large Language Models on Harmfulness",
    author = "Chen, Zixin  and
      Lin, Hongzhan  and
      Li, Kaixin  and
      Luo, Ziyang  and
      Ye, Zhen  and
      Chen, Guang  and
      Huang, Zhiyong  and
      Ma, Jing",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.213/",
    pages = "4234--4253",
    ISBN = "979-8-89176-251-0",
    abstract = "The proliferation of multimodal memes in the social media era demands that multimodal Large Language Models (mLLMs) effectively understand meme harmfulness. Existing benchmarks for assessing mLLMs on harmful meme understanding rely on accuracy-based, model-agnostic evaluations using static datasets. These benchmarks are limited in their ability to provide up-to-date and thorough assessments, as online memes evolve dynamically. To address this, we propose AdamMeme, a flexible, agent-based evaluation framework that adaptively probes the reasoning capabilities of mLLMs in deciphering meme harmfulness. Through multi-agent collaboration, AdamMeme provides comprehensive evaluations by iteratively updating the meme data with challenging samples, thereby exposing specific limitations in how mLLMs interpret harmfulness. Extensive experiments show that our framework systematically reveals the varying performance of different target mLLMs, offering in-depth, fine-grained analyses of model-specific weaknesses. Our code is available at https://github.com/Lbotirx/AdamMeme."
}