# MOVER: Multimodal Optimal Transport with Volume-based Embedding Regularization

arXiv: https://arxiv.org/abs/2508.12149 · HTML: https://mlsanity.com/paper/2508.12149/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 35 references that are in the map.
- First posted on arXiv 16 Aug 2025.

## Builds on (most cited first, 20 of 20)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [ImageBind: One Embedding Space To Bind Them All](https://mlsanity.com/paper/2305.05665/) (2023)
- [VGGSound: A Large-scale Audio-Visual Dataset](https://mlsanity.com/paper/2004.14368/) (2020)
- [VATEX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language Research](https://mlsanity.com/paper/1904.03493/) (2019)
- [BEATs: Audio Pre-Training with Acoustic Tokenizers](https://mlsanity.com/paper/2212.09058/) (2022)
- [LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment](https://mlsanity.com/paper/2310.01852/) (2023)
- [Clotho: An Audio Captioning Dataset](https://mlsanity.com/paper/1910.09387/) (2019)
- [A Short Note on the Kinetics-700 Human Action Dataset](https://mlsanity.com/paper/1907.06987/) (2019)
- [Multimodal Foundation Models: From Specialists to General-Purpose Assistants](https://mlsanity.com/paper/2309.10020/) (2023)
- [VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks](https://mlsanity.com/paper/2410.05160/) (2024)
- [VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset](https://mlsanity.com/paper/2305.18500/) (2023)
- [MM-Embed: Universal Multimodal Retrieval with Multimodal LLMs](https://mlsanity.com/paper/2411.02571/) (2024)
- [Multimodal Representation Learning by Alternating Unimodal Adaptation](https://mlsanity.com/paper/2311.10707/) (2023)
- [Audio Retrieval with Natural Language Queries: A Benchmark Study](https://mlsanity.com/paper/2112.09418/) (2021)
- [Gramian Multimodal Representation Learning and Alignment](https://mlsanity.com/paper/2412.11959/) (2024)
- [Recent Advances of Multimodal Continual Learning: A Comprehensive Survey](https://mlsanity.com/paper/2410.05352/) (2024)
- [Understanding the Emergence of Multimodal Representation Alignment](https://mlsanity.com/paper/2502.16282/) (2025)
- [Heterogeneous Contrastive Learning for Foundation Models and Beyond](https://mlsanity.com/paper/2404.00225/) (2024)
- [DecAlign: Hierarchical Cross-Modal Alignment for Decoupled Multimodal Representation Learning](https://mlsanity.com/paper/2503.11892/) (2025)
- [On the Importance of Contrastive Loss in Multimodal Learning](https://mlsanity.com/paper/2304.03717/) (2023)

## Built on by (most cited first, 6 of 6)

- [Modular MeanFlow: Towards Stable and Scalable One-Step Generative Modeling](https://mlsanity.com/paper/2508.17426/) (2025)
- [Gradient Shaping Beyond Clipping: A Functional Perspective on Update Magnitude Control](https://mlsanity.com/paper/2510.01578/) (2025)
- [ReSSFormer: A Recursive Sparse Structured Transformer for Scalable and Long-Context Reasoning](https://mlsanity.com/paper/2510.01585/) (2025)
- [Generalized Radius and Integrated Codebook Transforms for Differentiable Vector Quantization](https://mlsanity.com/paper/2602.01140/) (2026)
- [Hypergraph-Regularized Gramian Volumes for Multimodal Retrieval](https://mlsanity.com/paper/2609.15320/) (2026)
- [Query-Conditioned Spherical Centroid Aggregation for Multimodal Retrieval](https://mlsanity.com/paper/2609.15335/) (2026)

## Abstract (arXiv)

Recent advances in multimodal learning have largely relied on pairwise contrastive objectives to align different modalities, such as text, video, and audio, in a shared embedding space. While effective in bi-modal setups, these approaches struggle to generalize across multiple modalities and often lack semantic structure in high-dimensional spaces. In this paper, we propose MOVER, a novel framework that combines optimal transport-based soft alignment with volume-based geometric regularization to build semantically aligned and structured multimodal representations. By integrating a transport-guided matching mechanism with a geometric volume minimization objective (GAVE), MOVER encourages consistent alignment across all modalities in a modality-agnostic manner. Experiments on text-video-audio retrieval tasks demonstrate that MOVER significantly outperforms prior state-of-the-art methods in both zero-shot and finetuned settings. Additional analysis shows improved generalization to unseen modality combinations and stronger structural consistency in the learned embedding space.