@inproceedings{gonzalez-etal-2026-multimodal,
    title = "Multimodal Safety Evaluation in Generative Agent Social Simulations",
    author = "Gonzalez, Alhim Adonai Vera  and
      Hinojosa, Carlos  and
      Sanchez, Karen  and
      Hamid, Haidar Bin  and
      Kim, Donghoon  and
      Ghanem, Bernard",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1915/",
    pages = "41295--41310",
    ISBN = "979-8-89176-390-6",
    abstract = "Can generative agents be trusted in multimodal environments? Despite recent advances, agents remain limited in their ability to reason about safety, coherence, and trust across modalities. We introduce a reproducible simulation framework to evaluate generative agents in three aspects: (1) safety improvement over time via iterative plan revision in multimodal scenarios; (2) detection of unsafe activities across social contexts; and (3) social dynamics, measured through interaction and acceptance rates. These multimodal agents are evaluated using metrics that quantify plan revisions and unsafe-to-safe conversions. Experiments show that while agents detect direct multimodal contradictions, they often fail to align local revisions with global safety, achieving only a 55{\%} success rate in correcting unsafe plans. We release a dataset of 1,000 multimodal plans, yielding more than 600,000 simulation steps. Notably, 45{\%} of unsafe actions are accepted when paired with misleading visual cues, revealing a strong tendency to overtrust visual content. Code is available at https://github.com/AdonaiVera/X-CASE"
}