You Only Need Less Attention at Each Stage in Vision Transformers

TL;DR

This paper introduces LaViT, a vision transformer that reduces computational costs by computing attention only at certain stages and using attention transformations, achieving efficiency without sacrificing performance.

Contribution

The paper proposes a novel ViT architecture that computes less attention at each stage and uses attention transformations, addressing efficiency and saturation issues.

Findings

Achieves superior efficiency over traditional ViTs.

Performs well across classification, detection, and segmentation tasks.

Reduces computational complexity while maintaining high accuracy.

Abstract

The advent of Vision Transformers (ViTs) marks a substantial paradigm shift in the realm of computer vision. ViTs capture the global information of images through self-attention modules, which perform dot product computations among patchified image tokens. While self-attention modules empower ViTs to capture long-range dependencies, the computational complexity grows quadratically with the number of tokens, which is a major hindrance to the practical application of ViTs. Moreover, the self-attention mechanism in deep ViTs is also susceptible to the attention saturation issue. Accordingly, we argue against the necessity of computing the attention scores in every layer, and we propose the Less-Attention Vision Transformer (LaViT), which computes only a few attention operations at each stage and calculates the subsequent feature alignments in other layers via attention transformations that leverage the previously calculated attention scores. This novel approach can mitigate two primary issues plaguing traditional self-attention modules: the heavy computational burden and attention saturation. Our proposed architecture offers superior efficiency and ease of implementation, merely requiring matrix multiplications that are highly optimized in contemporary deep learning frameworks. Moreover, our architecture demonstrates exceptional performance across various vision tasks including classification, detection and segmentation.