• S2-MLP V1&V2 : Spatial-Shift MLP Architecture for Vision
• FishNet: A Versatile Backbone for Image, Region, and Pixel Level Prediction
• Attention Augmented Convolutional Networks
• Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
• LambdaNetworks: Modeling long-range Interactions without Attention
• Involution: Inverting the Inherence of Convolution for Visual Recognition
• ConvMLP: Hierarchical Convolutional MLPs for Vision
• Sparse-MLP: A Fully-MLP Architecture with Conditional Computation
• Hire-MLP: Vision MLP via Hierarchical Rearrangement
• RaftMLP: Do MLP-based Models Dream of Winning Over Computer Vision?
• CycleMLP: A MLP-like Architecture for Dense Prediction
• Vision Transformers with Hierarchical Attention
• X-volution: On the Unification of Convolution and Self-attention
• On the Integration of Self-Attention and Convolution
• DynaMixer: A Vision MLP Architecture with Dynamic Mixing
• ELSA: Enhanced Local Self-Attention for Vision Transformer
• Container: Context Aggregation Network
• Neighborhood Attention Transformer
• EdgeFormer: Improving Light-weight ConvNets by Learning from Vision Transformers
• TRT-ViT: TensorRT-oriented Vision Transformer
• Fast Vision Transformers with HiLo Attention
• ActiveMLP: An MLP-like Architecture with Active Token Mixer