AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

General-purpose, long-context autoregressive modeling with Perceiver AR

Curtis Hawthorne, Andrew Jaegle, Cătălina Cangea, Sebastian Borgeaud, Charlie Nash, Mateusz Malinowski, Sander Dieleman, Oriol Vinyals, Matthew Botvinick, Ian Simon, Hannah Sheahan, Neil Zeghidour and 3 more

ICML 2022 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models · 34 citations (OpenAlex)

This paper tried a common change: Larger batch size

Increase batch size from 64 to 128 while using 1024 latents · Copy task, context length 8192 · test accuracy: <1% -> 100% Improved

What 785 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
Use Perceiver AR for long-context autoregressive modelingImageNet 64×64 density estimation · ImageNet 64×64 valbits/dim: 3.40 versus 3.42Improved
Use Perceiver AR with 2048 context instead of Transformer-XLPG-19 language modeling · PG-19 testperplexity: 36.3 -> 28.9Improved
Extend context to 16384 in Perceiver ARBooks language modeling · Books test, compute-matchedperplexity: 14.276 -> 13.749Improved
Increase evaluation latents from 1024 to 1536 (ablation)Perceiver AR · ImageNet 64×64 valbits/dim: 3.4025 -> 3.3986Improved
Increase evaluation latents from 1024 to 4096 (ablation)Perceiver AR · ImageNet 64×64 valbits/dim: 3.4025 -> 3.5569Hurt
Reduce evaluation latents from 1024 to 16 (ablation)Perceiver AR · ImageNet 64×64 valbits/dim: 3.4025 -> 3.5664Hurt
Increase input context from 1024 to 12289 (ablation)Perceiver AR with R→G→B ordering · ImageNet 64×64 valbits/dim: 4.63 -> 3.53Improved
Increase input context from 1024 to 12289 (ablation)Perceiver AR with standard ordering · ImageNet 64×64 valbits/dim: 3.55 -> 3.54No clear effect
Increase context from 1024 to 16384 (ablation)Perceiver AR · Books evalperplexity: 14.88 -> 14.56Improved
Increase context from 1024 to 2048 (ablation)Perceiver AR · Wikitext-103 testperplexity: 18.52 -> 18.35Improved
Increase context from 2048 to 4096 (ablation)Perceiver AR · Wikitext-103 testperplexity: 18.35 -> 18.25Improved
Increase context from 2048 to 8192 (ablation)Perceiver AR · Wikitext-103 testperplexity: 18.35 -> 18.37Hurt

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