@inproceedings{wu-etal-2026-tags,
    title = "{TAGS}: A Test-Time Generalist{--}Specialist Framework with Retrieval-Augmented Reasoning and Verification",
    author = "Wu, Jianghao  and
      Tang, Feilong  and
      Li, Yulong  and
      Hu, Ming  and
      Xue, Haochen  and
      Jameel, Shoaib  and
      Ge, Zongyuan  and
      Xie, Yutong  and
      Razzak, Imran",
    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.757/",
    pages = "15428--15445",
    ISBN = "979-8-89176-395-1",
    abstract = "Recent advances such as Chain-of-Thought prompting have significantly improved large language models (LLMs) in zero-shot medical reasoning. However, prompting-based methods often remain shallow and unstable, while fine-tuned medical LLMs suffer from poor generalization under distribution shifts and limited adaptability to unseen clinical scenarios. To address these limitations, we present TAGS, a test-time framework that combines a broadly capable generalist with a domain-specific specialist to offer complementary perspectives without any model fine-tuning or parameter updates. To support this generalist{--}specialist reasoning process, we introduce two auxiliary modules: a hierarchical retrieval mechanism that provides multi-scale exemplars by selecting examples based on both semantic and rationale-level similarity, and a reliability scorer that evaluates reasoning consistency to guide final answer aggregation. TAGS achieves strong performance across nine MedQA benchmarks, boosting GPT-4o accuracy by 13.8{\%}, DeepSeek-R1 by 16.8{\%}, and improving a vanilla 7B model from 14.1{\%} to 23.9{\%}. These results surpass several fine-tuned medical LLMs, without any parameter updates."
}