Methods › Natural Language Processing › Transformers › Sparse Transformer

Sparse Transformer

47 papers tagged archive 2025-07-28

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

A Sparse Transformer is a Transformer based architecture which utilises sparse factorizations of the attention matrix to reduce time/memory to O(n √(n)). Other changes to the Transformer architecture include: (a) a restructured residual block and weight initialization, (b) A set of sparse attention kernels which efficiently compute subsets of the attention matrix, (c) recomputation of attention weights during the backwards pass to reduce memory usage

Source: Generating Long Sequences with Sparse Transformers

Papers archive 2025-07-28

30 shown of 47, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 80 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Language Modelling6
Decoder5
Language Modeling4
Mixture-of-Experts4
Text Classification4
text-classification4
Diversity3
GPU3
Image Restoration3
Machine Translation3
Object3
Object Detection3
Question Answering3
Semantic Segmentation3
Translation3
object-detection3
CPU2
Clustering2
Document Summarization2
Image Captioning2

Usage over time archive 2025-07-28

Papers per year tagged with Sparse Transformer: 2019 to 2025, peak 12 12 0 2019: 3 papers 2019 2020: 5 papers 2020 2021: 6 papers 2021 2022: 10 papers 2022 2023: 8 papers 2023 2024: 12 papers 2024 2025: 3 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (47 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Transformers

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections