Skip to content
conftrace_

softmax nonlinearity

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

selective self-attention (1) context sparsity (1) temperature scaling (1) contextual sparsity (1) selective attention (1) language modeling (1) attention mechanism (1) transformer architecture (1)

Papers

Selective Attention: Enhancing Transformer through Principled Context Control NeurIPS 2024