@inproceedings{choi-etal-2025-zero,
    title = "Zero-shot Multimodal Document Retrieval via Cross-modal Question Generation",
    author = "Choi, Yejin  and
      Park, Jaewoo  and
      Yoon, Janghan  and
      Kim, Saejin  and
      Jeon, Jaehyun  and
      Yu, Youngjae",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.1324/",
    pages = "26079--26094",
    ISBN = "979-8-89176-332-6",
    abstract = "Rapid advances in Multimodal Large Language Models (MLLMs) have extended information retrieval beyond text, enabling access to complex real-world documents that combine both textual and visual content. However, most documents are private, either owned by individuals or confined within corporate silos, and current retrievers struggle when faced with unseen domains or languages. To address this gap, we introduce PREMIR, a simple yet effective framework that leverages the broad knowledge of an MLLM to generate cross-modal pre-questions (preQs) before retrieval. Unlike earlier multimodal retrievers that embed entire documents as a single vector, PREMIR leverages preQs, decomposed from documents into finer token-level representations across modalities, enabling richer contextual understanding. Experiments show that PREMIR achieves state-of-the-art performance on out-of-distribution benchmarks, including closed-domain and multilingual settings, outperforming strong baselines across all metrics. We confirm the contribution of each component through in-depth ablation studies, and qualitative analyses of the generated preQs further highlight the framework{'}s robustness in real-world settings."
}