Skip to yearly menu bar Skip to main content


Poster

Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers

Sanghyeok Lee · Joonmyung Choi · Hyunwoo J. Kim

Arch 4A-E Poster #109
[ ] [ Project Page ] [ Paper PDF ]
[ Poster
Thu 20 Jun 5 p.m. PDT — 6:30 p.m. PDT

Abstract:

Vision Transformer (ViT) has emerged as a prominent backbone for computer vision. For more efficient ViTs, recent works lessen the quadratic cost of the self-attention layer by pruning or fusing the redundant tokens. However, these works faced the speed-accuracy trade-off caused by the loss of information. Here, we argue that token fusion needs to consider diverse relations between tokens to minimize information loss. In this paper, we propose a Multi-criteria Token Fusion, that gradually fuses the tokens based on multi-criteria (\ie, similarity, informativeness, and size of fused tokens). Further, we utilize the one-step-ahead attention, which is the improved approach to capture the informativeness of the tokens.By training the model equipped with MCTF using a token reduction consistency, we achieve the best speed-accuracy trade-off in the image classification (ImageNet1K). Experimental results prove that MCTF consistently surpasses the previous reduction methods with and without training. Specifically, DeiT-T and DeiT-S with MCTF reduce FLOPs by about 44% while improving the performance (+0.5%, and +0.3%) over the base model, respectively. We also demonstrate the applicability of MCTF in various Vision Transformers (e.g., T2T-ViT, LV-ViT), achieving at least 31% speedup without performance degradation.

Chat is not available.