@inproceedings{zhang-etal-2026-d2,
    title = "D$^2$-{RAG}: Dual-Decision Retrieval-Augmented Generation via Multi-Dimensional Uncertainty and Utility-Aware Decoding",
    author = "Zhang, Jinshuo  and
      Zhou, Xiaoding  and
      Zhang, Weiyu  and
      Chen, Guoqiang  and
      Lian, Ying  and
      Meng, Xiaoyang  and
      Chen, Yonghe  and
      Guan, Hongjiao  and
      Si, Jiasheng  and
      Lu, Wenpeng",
    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.754/",
    pages = "15355--15382",
    ISBN = "979-8-89176-395-1",
    abstract = "Retrieval-Augmented Generation (RAG) mitigates hallucinations in large language models by incorporating external knowledge. However, retrieval does not always return relevant documents and may return noisy ones. Indiscriminately retrieving and utilizing this external knowledge can interfere with the model{'}s originally correct reasoning. In this work, we propose Dual-Decision Retrieval-Augmented Generation (D$^2$-RAG), which integrates multi-dimensional uncertainty estimation to decide whether to retrieve and employs adaptive contrastive decoding to handle retrieved contexts of varying quality. Specifically, we first integrate uncertainty estimation scores that assess model uncertainty from multiple perspectives, construct them into a comprehensive feature vector, and train a lightweight retrieval decision model to accurately identify the model{'}s knowledge boundaries and determine whether to retrieve. Subsequently, we dynamically adjust the contrastive decoding strategy based on the utility of retrieved contexts to enhance the utilization of relevant contexts while suppressing interference from noisy contexts. Extensive experiments on four medical question-answering datasets demonstrate that D$^2$-RAG significantly outperforms baselines, enabling retrieval-augmented Llama3.1-8B to surpass non-retrieval-augmented Llama3.1-70B on the MedMCQA dataset. The source code is available on https://github.com/zakelawen/d{--}rag."
}