SLAB: Efficient Transformers with Simplified Linear Attention and Progressive Re-parameterized Batch Normalization
This work addresses the deployment challenge of Transformers for resource-constrained devices in computer vision and language modeling, offering an incremental improvement in efficiency and performance.
The paper tackles the computational inefficiency of Transformers on resource-constrained devices by proposing SLAB, which combines a progressive re-parameterized BatchNorm (PRepBN) to replace LayerNorm and a simplified linear attention (SLA) module. It achieves 83.6% top-1 accuracy on ImageNet-1K with 16.2ms latency, reducing latency by 2.4ms while slightly improving accuracy compared to a baseline.
Transformers have become foundational architectures for both natural language and computer vision tasks. However, the high computational cost makes it quite challenging to deploy on resource-constraint devices. This paper investigates the computational bottleneck modules of efficient transformer, i.e., normalization layers and attention modules. LayerNorm is commonly used in transformer architectures but is not computational friendly due to statistic calculation during inference. However, replacing LayerNorm with more efficient BatchNorm in transformer often leads to inferior performance and collapse in training. To address this problem, we propose a novel method named PRepBN to progressively replace LayerNorm with re-parameterized BatchNorm in training. Moreover, we propose a simplified linear attention (SLA) module that is simple yet effective to achieve strong performance. Extensive experiments on image classification as well as object detection demonstrate the effectiveness of our proposed method. For example, our SLAB-Swin obtains $83.6\%$ top-1 accuracy on ImageNet-1K with $16.2$ms latency, which is $2.4$ms less than that of Flatten-Swin with $0.1\%$ higher accuracy. We also evaluated our method for language modeling task and obtain comparable performance and lower latency.Codes are publicly available at https://github.com/xinghaochen/SLAB and https://github.com/mindspore-lab/models/tree/master/research/huawei-noah/SLAB.