ToSA: Token Selective Attention for Efficient Vision Transformers
This addresses efficiency issues for researchers and practitioners using vision transformers in resource-constrained settings, though it is incremental as it builds on existing transformer architectures.
The paper tackles the problem of high computational costs in vision transformers by proposing ToSA, a token selective attention method that reduces computation by skipping less important tokens in transformer layers, achieving significant computation reduction while maintaining accuracy on ImageNet classification and similar performance on monocular depth estimation with a lighter backbone.
In this paper, we propose a novel token selective attention approach, ToSA, which can identify tokens that need to be attended as well as those that can skip a transformer layer. More specifically, a token selector parses the current attention maps and predicts the attention maps for the next layer, which are then used to select the important tokens that should participate in the attention operation. The remaining tokens simply bypass the next layer and are concatenated with the attended ones to re-form a complete set of tokens. In this way, we reduce the quadratic computation and memory costs as fewer tokens participate in self-attention while maintaining the features for all the image patches throughout the network, which allows it to be used for dense prediction tasks. Our experiments show that by applying ToSA, we can significantly reduce computation costs while maintaining accuracy on the ImageNet classification benchmark. Furthermore, we evaluate on the dense prediction task of monocular depth estimation on NYU Depth V2, and show that we can achieve similar depth prediction accuracy using a considerably lighter backbone with ToSA.