Pay Attention to MLPs

Abstract

Transformers have become one of the most important architectural innovations in deep learning and have enabled many breakthroughs over the past few years. Here we propose a simple network architecture, gMLP, based on MLPs with gating, and show that it can perform as well as Transformers in key language and vision applications. Our comparisons show that self-attention is not critical for Vision Transformers, as gMLP can achieve the same accuracy. For BERT, our model achieves parity with Transformers on pretraining perplexity and is better on some downstream NLP tasks. On finetuning tasks where gMLP performs worse, making the gMLP model substantially larger can close the gap with Transformers. In general, our experiments show that gMLP can scale as well as Transformers over increased data and compute.

References in corpus (7)

• Scaling Laws for Neural Language Models
• MLP-Mixer: An all-MLP Architecture for Vision
• Pay Less Attention with Lightweight and Dynamic Convolutions
• High-Performance Large-Scale Image Recognition Without Normalization
• CvT: Introducing Convolutions to Vision Transformers
• LambdaNetworks: Modeling Long-Range Interactions Without Attention
• GLU Variants Improve Transformer

Cited by in corpus (9)

• S${}^2$-MLPv2: Improved Spatial-Shift MLP Architecture for Vision
• S${}^2$-MLP: Spatial-Shift MLP Architecture for Vision
• ConvMLP: Hierarchical Convolutional MLPs for Vision
• Vision Permutator: A Permutable MLP-Like Architecture for Visual Recognition
• Global Filter Networks for Image Classification
• Visual Transformer with Statistical Test for COVID-19 Classification
• Rethinking Token-Mixing MLP for MLP-based Vision Backbone
• MOI-Mixer: Improving MLP-Mixer with Multi Order Interactions in Sequential Recommendation
• Convolutional Gated MLP: Combining Convolutions & gMLP