AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Perceiver: General Perception with Iterative Attention

Andrew Jaegle, Felix Gimeno, Andy Brock, Oriol Vinyals, Andrew Zisserman, Joao Carreira

ICML 2021 · Architectures and efficiency › Attention and transformers · approach: Theory of self-attention dynamics · 125 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 Fourier features with learned position encodings (ablation)Perceiver · ImageNet validationtop-1 accuracy: 78.0 -> 70.9Hurt
interleave 8 cross-attends through the network rather than placing them at the start (ablation)Perceiver · ImageNet validationtop-1 accuracy: 73.7 -> 78.0Improved
increase cross-attends from 4 to 8 while interleaving them (ablation)Perceiver · ImageNet validationtop-1 accuracy: 76.5 -> 78.0Improved
remove weight sharing (ablation)Perceiver · ImageNet validationtop-1 accuracy: 78.0 -> 72.9Hurt
use weight sharing (ablation)Perceiver · ImageNet traintop-1 accuracy: 87.7 -> 79.5Hurt
use weight sharing (ablation)Perceiver · ImageNet validationtop-1 accuracy: 72.9 -> 78.0Improved
use crop-relative rather than raw-image position coordinates (ablation)Perceiver · ImageNettop-1 accuracyImproved
apply video dropout during audiovisual training (ablation)Perceiver · AudioSet audio + videomAP: 39.9% vs. 43.2%; 39.7% vs. 43.5%Improved
turn off SpecAugment and drop the spectrogram modality with 10% probability (ablation)Perceiver mel-spectrogram model · AudioSetmAP: 44.2Improved
use temporal unrolling, processing one frame per cross-attend (ablation)Perceiver · AudioSet videoperformanceImproved
use temporal unrolling, processing one frame per cross-attend (ablation)Perceiver · AudioSet audioperformanceHurt
add random per-point translation and rotation augmentation (ablation)Perceiver · ModelNet40accuracyNo clear effect

Later papers that compared with this method

1 later accepted paper reports Perceiver as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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