Multimodal information retrieval (MIR) faces inherent challenges due to the heterogeneity of data sources and the complexity of cross-modal alignment. While previous studies have identified modal gaps in feature spaces, a systematic approach to address these challenges remains unexplored. In this work, we introduce UNITE, a universal framework that tackles these challenges through two critical yet underexplored aspects: data curation and modality-aware training configurations. Our work provides the first comprehensive analysis of how modality-specific data properties influence downstream task performance across diverse scenarios. Moreover, we propose Modal-Aware Masked Contrastive Learning (MAMCL) to mitigate the competitive relationships among the instances of different modalities. Our framework achieves state-of-the-art results on multiple multimodal retrieval benchmarks, outperforming existing methods by notable margins. Through extensive experiments, we demonstrate that strategic modality curation and tailored training protocols are pivotal for robust cross-modal representation learning. This work not only advances MIR performance but also provides a foundational blueprint for future research in multimodal systems. Our project is available at https://friedrichor.github.io/projects/UNITE.
@article{arxiv.2505.19650,
title = {Modality Curation: Building Universal Embeddings for Advanced Multimodal Information Retrieval},
author = {Fanheng Kong and Jingyuan Zhang and Yahui Liu and Hongzhi Zhang and Shi Feng and Xiaocui Yang and Daling Wang and Yu Tian and Victoria W. and Fuzheng Zhang and Guorui Zhou},
journal= {arXiv preprint arXiv:2505.19650},
year = {2025}
}