@inproceedings{qiu-etal-2025-entropy,
    title = "Entropy-Based Decoding for Retrieval-Augmented Large Language Models",
    author = "Qiu, Zexuan  and
      Ou, Zijing  and
      Wu, Bin  and
      Li, Jingjing  and
      Liu, Aiwei  and
      King, Irwin",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/moar-dois/2025.naacl-long.236/",
    doi = "10.18653/v1/2025.naacl-long.236",
    pages = "4616--4627",
    ISBN = "979-8-89176-189-6",
    abstract = "Augmenting Large Language Models (LLMs) with retrieved external knowledge has proven effective in improving the factual accuracy of generated responses. Despite their success, retrieval-augmented LLMs still face the distractibility issue, where the generated responses are negatively influenced by noise from both external and internal knowledge sources. In this paper, we introduce a novel, training-free decoding method guided by entropy considerations to mitigate this issue. Our approach utilizes entropy-based document-parallel ensemble decoding to prioritize low-entropy distributions from retrieved documents, thereby enhancing the extraction of relevant information of context. Additionally, it incorporates a contrastive decoding mechanism that contrasts the obtained low-entropy ensemble distribution with the high-entropy distribution derived from the model{'}s internal knowledge across layers, which ensures a greater emphasis on reliable external information. Extensive experiments on open-domain question answering datasets demonstrate the superiority of our method."
}