Methods › General › Regularization › Attention Dropout

Attention Dropout

introduced 2018 10,892 papers tagged archive 2025-07-28

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

Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the softmax in the attention equation. For example, for scaled-dot product attention, we would drop elements from the first term:

Attention(Q, K, V) = softmax(QKᵀ/(√(dₖ)))V

See Code · huggingface/transformers

Papers archive 2025-07-28

30 shown of 10,892, 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 1,501 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 Modelling1,902
Language Modeling1,489
Retrieval1,433
RAG1,308
Retrieval-augmented Generation1,122
Question Answering1,070
Sentence920
Large Language Model501
Sentiment Analysis470
Text Generation470
Text Classification459
text-classification409
Transfer Learning393
Natural Language Understanding311
Information Retrieval302
Classification299
Decoder295
Translation294
Word Embeddings280
Named Entity Recognition273

Usage over time archive 2025-07-28

Papers per year tagged with Attention Dropout: 2015 to 2025, peak 2,727 2,727 0 2015: 1 paper 2015 2016: 0 papers 2016 2017: 0 papers 2017 2018: 10 papers 2018 2019: 609 papers 2019 2020: 1367 papers 2020 2021: 1623 papers 2021 2022: 1298 papers 2022 2023: 2079 papers 2023 2024: 2727 papers 2024 2025: 1178 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (10,892 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

Regularization

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