English

Efficient Multimodal Fusion via Interactive Prompting

Computer Vision and Pattern Recognition 2023-05-16 v2

Abstract

Large-scale pre-training has brought unimodal fields such as computer vision and natural language processing to a new era. Following this trend, the size of multi-modal learning models constantly increases, leading to an urgent need to reduce the massive computational cost of finetuning these models for downstream tasks. In this paper, we propose an efficient and flexible multimodal fusion method, namely PMF, tailored for fusing unimodally pre-trained transformers. Specifically, we first present a modular multimodal fusion framework that exhibits high flexibility and facilitates mutual interactions among different modalities. In addition, we disentangle vanilla prompts into three types in order to learn different optimizing objectives for multimodal learning. It is also worth noting that we propose to add prompt vectors only on the deep layers of the unimodal transformers, thus significantly reducing the training memory usage. Experiment results show that our proposed method achieves comparable performance to several other multimodal finetuning methods with less than 3% trainable parameters and up to 66% saving of training memory usage.

Keywords

Cite

@article{arxiv.2304.06306,
  title  = {Efficient Multimodal Fusion via Interactive Prompting},
  author = {Yaowei Li and Ruijie Quan and Linchao Zhu and Yi Yang},
  journal= {arXiv preprint arXiv:2304.06306},
  year   = {2023}
}

Comments

Camera-ready version for CVPR2023

R2 v1 2026-06-28T10:03:46.111Z