cs.CVNov 12, 2026

Model Distribution-Aware Multimodal Dataset Distillation

Authors: Jiajun Shen, Yonglin Wu, Ruonan Yu, Xinchao Wang

Organizations: National University of Singapore

Abstract

Multimodal dataset distillation (MDD) aims to reduce the cost of vision-language pre-training by synthesizing compact image-text datasets. Although recent feature-matching methods have improved distillation efficiency, synthetic data may become specialized to the model representations encountered during optimization, limiting generalization to unseen model states. We propose Model Distribution-Aware Multimodal Dataset Distillation (MDA-DD), which leverages supervision from a dynamic staggered queue of models at different training stages. We further introduce asymmetric dual projection for complementary alignment in visual and linguistic spaces and an uncentered second-order moment objective that retains mean-related cross-modal information. Experiments on Flickr30K and MS-COCO demonstrate consistent retrieval improvements over existing methods across synthetic-data budgets. With only 100 synthetic pairs, MDA-DD improves IR@1 and TR@1 on Flickr30K by 12.9 and 18.2 percentage points, respectively, over the strongest baseline at the same budget, outperforming all compared methods using 500 pairs on both datasets across all six retrieval metrics.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Multimodal Distribution Matching for Vision-Language Dataset Distillation

    May 22, 2026Jongoh Jeong, Hoyong Kwon, Minseok Kim +1Cross-Modal LearningCross-Modal Alignment

  2. Multimodal Dataset Distillation Made Simple by Prototype-Guided Data Synthesis

    Feb 23, 2026Junhyeok Choi, Sangwoo Mo, Minwoo ChaeDataset DistillationMultimodal Dataset

  3. Multimodal Dataset Distillation via Phased Teacher Models

    Mar 26, 2026Shengbin Guo, Hang Zhao, Senqiao Yang +5Dataset DistillationKnowledge Distillation