@inproceedings{chen-etal-2025-vlm,
    title = "{VLM} Is a Strong Reranker: Advancing Multimodal Retrieval-augmented Generation via Knowledge-enhanced Reranking and Noise-injected Training",
    author = "Chen, Zhanpeng  and
      Xu, Chengjin  and
      Qi, Yiyan  and
      Jiang, Xuhui  and
      Guo, Jian",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.432/",
    doi = "10.18653/v1/2025.findings-emnlp.432",
    pages = "8140--8158",
    ISBN = "979-8-89176-335-7",
    abstract = "Vision-language Models (VLMs) have demonstrated remarkable capabilities in processing and generating content across multiple data modalities. However, a significant drawback of VLMs is their reliance on static training data, leading to outdated information and limited contextual awareness. This static nature hampers their ability to provide accurate and up-to-date responses, particularly in dynamic or rapidly evolving contexts. To address these limitations, we propose RagVL, a novel framework with knowledge-enhanced reranking and noise-injected training. We instruction-tune the VLM with a simple yet effective instruction template to induce its ranking ability and serve it as a reranker to precisely filter the top-k retrieved images. For generation, we inject visual noise during training at the data and token levels to enhance the generator{'}s robustness. Extensive experiments on four datasets verify the effectiveness of our method. Code and models are available at https://anonymous.4open.science/r/RagVL-F694."
}