模態策展：構建通用嵌入以實現先進的多模態信息檢索

Modality Curation: Building Universal Embeddings for Advanced Multimodal Information Retrieval

May 26, 2025

作者: Fanheng Kong, Jingyuan Zhang, Yahui Liu, Hongzhi Zhang, Shi Feng, Xiaocui Yang, Daling Wang, Yu Tian, Victoria W., Fuzheng Zhang, Guorui Zhou

cs.AI

摘要

多模態信息檢索（MIR）由於數據源的異質性和跨模態對齊的複雜性，面臨著固有的挑戰。儘管先前的研究已經識別出特徵空間中的模態差距，但解決這些挑戰的系統性方法仍未被探索。在本研究中，我們引入了UNITE，這是一個通用框架，通過兩個關鍵但尚未充分探索的方面來應對這些挑戰：數據策展和模態感知的訓練配置。我們的工作首次全面分析了模態特定數據屬性如何影響多樣化場景中的下游任務性能。此外，我們提出了模態感知掩碼對比學習（MAMCL）來緩解不同模態實例之間的競爭關係。我們的框架在多個多模態檢索基準上取得了最先進的成果，顯著超越了現有方法。通過大量實驗，我們證明了策略性的模態策展和定制的訓練協議對於穩健的跨模態表示學習至關重要。這項工作不僅提升了MIR的性能，還為未來多模態系統的研究提供了基礎藍圖。我們的項目可在https://friedrichor.github.io/projects/UNITE獲取。

English

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.