Principled Multimodal Representation Learning
Liu, Xiaohao, Xia, Xiaobo, Ng, See-Kiong, Chua, Tat-Seng
–arXiv.org Artificial Intelligence
Multimodal representation learning seeks to create a unified representation space by integrating diverse data modalities to improve multimodal understanding. Traditional methods often depend on pairwise contrastive learning, which relies on a predefined anchor modality, restricting alignment across all modalities. Recent advances have investigated the simultaneous alignment of multiple modalities, yet several challenges remain, such as limitations imposed by fixed anchor points and instability arising from optimizing the product of singular values. To address the challenges, in this paper, we propose Principled Multimodal Representation Learning (PMRL), a novel framework that achieves simultaneous alignment of multiple modalities without anchor dependency in a more stable manner. Specifically, grounded in the theoretical insight that full alignment corresponds to a rank-1 Gram matrix, PMRL optimizes the dominant singular value of the representation matrix to align modalities along a shared leading direction. We propose a softmax-based loss function that treats singular values as logits to prioritize the largest singular value. Besides, instance-wise contrastive regularization on the leading eigenvectors maintains inter-instance separability and prevents representation collapse. Extensive experiments across diverse tasks demonstrate PMRL's superiority compared to baseline methods. The source code will be publicly available.
arXiv.org Artificial Intelligence
Oct-28-2025
- Genre:
- Research Report > New Finding (0.46)
- Industry:
- Health & Medicine
- Therapeutic Area > Neurology (1.00)
- Health Care Technology (0.68)
- Health & Medicine
- Technology:
- Information Technology
- Data Science (0.93)
- Artificial Intelligence
- Representation & Reasoning (1.00)
- Natural Language (1.00)
- Vision (0.92)
- Machine Learning
- Statistical Learning (1.00)
- Neural Networks > Deep Learning (0.46)
- Performance Analysis > Accuracy (0.46)
- Information Technology