Text-to-image generation increasingly demands access to domain-specific, fine-grained, and rapidly evolving knowledge that pretrained models cannot fully capture, necessitating the integration of retrieval methods. Existing Retrieval-Augmented Generation (RAG) methods attempt to address this by retrieving globally relevant images, but they fail when no single image contains all desired elements from a complex user query. We propose Cross-modal RAG, a novel framework that decomposes both queries and images into sub-dimensional components, enabling subquery-aware retrieval and generation. Our method introduces a hybrid retrieval strategy - combining a sub-dimensional sparse retriever with a dense retriever - to identify a Pareto-optimal set of images, each contributing complementary aspects of the query. During generation, a multimodal large language model is guided to selectively condition on relevant visual features aligned to specific subqueries, ensuring subquery-aware image synthesis. Extensive experiments on MS-COCO, Flickr30K, WikiArt, CUB, and ImageNet-LT demonstrate that Cross-modal RAG significantly outperforms existing baselines in the retrieval and further contributes to generation quality, while maintaining high efficiency.
@article{arxiv.2505.21956,
title = {Cross-modal RAG: Sub-dimensional Text-to-Image Retrieval-Augmented Generation},
author = {Mengdan Zhu and Senhao Cheng and Guangji Bai and Yifei Zhang and Liang Zhao},
journal= {arXiv preprint arXiv:2505.21956},
year = {2025}
}