Dynamic Self-adaptive Multiscale Distillation from Pre-trained Multimodal Large Model for Efficient Cross-modal Retrieval | OpenReview

Dynamic Self-adaptive Multiscale Distillation from Pre-trained Multimodal Large Model for Efficient Cross-modal Retrieval

Open Webpage

Zhengyang Liang, Meiyu Liang, Wei Huang, Yawen Li, Wu Liu, Yingxia Shao, Kangkang Lu

Published: 27 Oct 2025, Last Modified: 21 Jan 2026CrossrefEveryoneRevisionsCC BY-SA 4.0

External IDs:doi:10.1145/3746027.3754844

Loading