AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Residual Shuffle-Exchange Networks for Fast Processing of Long Sequences

Andis Draguns, Emīls Ozoliņš, Agris Šostaks, Matīss Apinis, Karlis Freivalds

AAAI 2021 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace SE's gated Switch Unit with the residual Switch UnitShuffle-Exchange network · LAMBADA testaccuracy: 52.28 -> 54.34Improved
use RSE instead of the Translation-invariant netMusicNet · MusicNetAPS: 77.3 -> 78.02Improved
prepend 1 convolution instead of none (ablation)RSE · MusicNet, window 1024APS: 69.29 -> 69.57Improved
prepend 2 convolutions instead of none (ablation)RSE · MusicNet, window 1024APS: 69.29 -> 68.95No clear effect
prepend 3 convolutions instead of none (ablation)RSE · MusicNet, window 1024APS: 69.29 -> 67.39Hurt
add auxiliary note-prediction losses at regularly spaced sequence positions (ablation)RSE · MusicNetAPS: 76.84% -> 78.02%Improved
remove LayerNorm from the Residual Switch Unit (ablation)RSE · binary multiplication, length 128test errorHurt
replace GELU with ReLU in the Residual Switch Unit (ablation)RSE · binary multiplication, length 128test errorHurt
remove the residual connection from the Residual Switch Unit (ablation)RSE · binary multiplication, length 128test errorHurt
replace learnable residual scaling with a constant scale of 1 (ablation)RSE · binary multiplication, length 128test errorHurt
increase the Residual Switch Unit hidden layer size (ablation)RSE · binary multiplication, length 128test errorImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy