Method · Trajectory matching
AMD
Asynchronous Matching with Dynamic Sampling for Multimodal Dataset Distillation
Ding Qi, Jian Li, Shuguang Dou, Zifan Song, Junyao Gao, Yabiao Wang, Chengjie Wang, Cairong Zhao
ICLR 2026 · first public 2026-01
In one paragraph
Proposes AMD for image-text dataset distillation, performing asynchronous trajectory matching that decouples the starting points of image and text expert trajectories to accommodate their different learning speeds, and replacing random prototype initialization with a Semantics-Aware Prototype Mining module that clusters the feature space into representative prototypes; reported Recall@1/5/10 gains are 4.5, 9.6 and 10.9 points on Flickr30K at 200 distilled pairs.
Where it sits
- Trajectory matching (Surrogate matching)
- Setting: Image–text pairs (vision–language)
Design choices
| Labels | hard |
| Prior / networks used | trained-experts |
| Optimization regime | bilevel |
| What is stored | pixels, embedding-features |
| Largest scale evaluated | flickr30k-coco |
BibTeX (generated; prefer the venue's official entry)
@article{qi2026asynchronous,
title = {Asynchronous Matching with Dynamic Sampling for Multimodal Dataset Distillation},
author = {Ding Qi and Jian Li and Shuguang Dou and Zifan Song and Junyao Gao and Yabiao Wang and Chengjie Wang and Cairong Zhao},
journal = {ICLR 2026},
year = {2026}
}Nearby in Trajectory matching
PTM-ST — Multimodal Dataset Distillation via Phased Teacher Models
Shengbin Guo, Hang Zhao, Senqiao Yang et al. · ICLR 2026notableVision–languagepaper ↗code ↗
RepBlend — Beyond Modality Collapse: Representations Blending for Multimodal Dataset Distillation
Xin Zhang, Ziruo Zhang, Jiawei Du et al. · NeurIPS 2025notableVision–languagepaper ↗
MKDT — Dataset Distillation via Knowledge Distillation: Towards Efficient Self-Supervised Pre-Training of Deep Networks
Siddharth Joshi, Jiayi Ni, Baharan Mirzasoleiman · ICLR 2025notablePre-training & transferpaper ↗code ↗