@inproceedings{fu-etal-2025-exclusion,
    title = "Exclusion of Thought: Mitigating Cognitive Load in Large Language Models for Enhanced Reasoning in Multiple-Choice Tasks",
    author = "Fu, Qihang  and
      Qin, Yongbin  and
      Huang, Ruizhang  and
      Chen, Yanping  and
      Zhou, Yulin  and
      Long, Lintao",
    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.1051/",
    pages = "21673--21686",
    ISBN = "979-8-89176-251-0",
    abstract = "Multiple-choice questions (MCQs) are a widely used and vital assessment format for evaluating large language models (LLMs). This study reveals that LLMs are susceptible to ``cognitive load'' caused by distractor options in MCQs, leading to excessive attention to distractors and consequent vacillation between correct and incorrect options. To mitigate this cognitive burden, we introduce a novel reasoning prompt strategy, called EoT, which effectively reduces cognitive load by steering the model{'}s attention away from erroneous options. This enables the model to focus more effectively on reasonable answers. Additionally, by documenting the elimination process, EoT enhances the transparency and interpretability of the model{'}s reasoning. Experimental results demonstrate that EoT, as a plug-and-play approach, significantly reduces cognitive load and improves performance, showcasing its potential to enhance both the accuracy and interpretability of LLMs."
}