@inproceedings{liu-etal-2026-benchmarking,
    title = "Benchmarking Egocentric Clinical Intent Understanding Capability for Medical Multimodal Large Language Models",
    author = "Liu, Shaonan  and
      Yu, Guo  and
      Luo, Xiaoling  and
      Zheng, Shiyi  and
      Liu, Jie  and
      Chen, Wenting  and
      Shen, Linlin",
    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.1228/",
    pages = "26682--26697",
    ISBN = "979-8-89176-390-6",
    abstract = "Medical Multimodal Large Language Models (Med-MLLMs) require egocentric clinical intent understanding for real-world deployment, yet existing benchmarks fail to evaluate this critical capability. We introduce MedGaze-Bench, the first benchmark leveraging clinician gaze as a Cognitive Cursor to assess intent understanding across surgery, emergency simulation, and diagnostic interpretation. Our benchmark addresses three fundamental challenges: visual homogeneity of anatomical structures, strict temporal-causal dependencies in clinical workflows, and implicit adherence to safety protocols. We propose a Three-Dimensional Clinical Intent Framework evaluating: (1) Spatial Intent{---}discriminating precise targets amid visual noise, (2) Temporal Intent{---}inferring causal rationale through retrospective and prospective reasoning, and (3) Standard Intent{---}verifying protocol compliance through safety checks. Beyond accuracy metrics, we introduce Trap QA mechanisms to stress-test clinical reliability by penalizing hallucinations and cognitive sycophancy. Experiments reveal current MLLMs struggle with egocentric intent due to over-reliance on global features, leading to fabricated observations and uncritical acceptance of invalid instructions."
}