Skip to yearly menu bar Skip to main content


Group Equivariant Stand-Alone Self-Attention For Vision

David W. Romero · Jean-Baptiste Cordonnier


Keywords: [ group equivariance ] [ group equivariant self-attention ] [ group equivariant transformers ] [ transformers ] [ self-attention ]


We provide a general self-attention formulation to impose group equivariance to arbitrary symmetry groups. This is achieved by defining positional encodings that are invariant to the action of the group considered. Since the group acts on the positional encoding directly, group equivariant self-attention networks (GSA-Nets) are steerable by nature. Our experiments on vision benchmarks demonstrate consistent improvements of GSA-Nets over non-equivariant self-attention networks.

Chat is not available.