@inproceedings{zhang-etal-2026-beyond-prompt,
    title = "Beyond Prompt: Fine-grained Simulation of Cognitively Impaired Standardized Patients via Stochastic Steering",
    author = "Zhang, Weikang  and
      Zhu, Zimo  and
      Yang, Zhichuan  and
      Huang, Chen  and
      Lei, Wenqiang  and
      Ng, See-Kiong",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1401/",
    pages = "28098--28131",
    ISBN = "979-8-89176-395-1",
    abstract = "Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation of cognitively impaired patients. We innovatively capture domain-specific features by extracting steering vectors from contrastive pairs of instructions and responses. Furthermore, we introduce a Stochastic Token Modulation (STM) mechanism to regulate the intervention probability. STM enables precise control over impairment severity while mitigating the instability of conventional vector methods. Comprehensive experiments demonstrate that StsPatient significantly outperforms baselines in both clinical authenticity and severity controllability. Our code will be open-sourced upon acceptance."
}