Methods › General › Attention Mechanisms › SortCut Sinkhorn Attention

SortCut Sinkhorn Attention

1 paper tagged archive 2025-07-28

Introduced by Yi Tay et al. in Sparse Sinkhorn Attention

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

SortCut Sinkhorn Attention is a variant of Sparse Sinkhorn Attention where a post-sorting truncation of the input sequence is performed, essentially performing a hard top-k operation on the input sequence blocks within the computational graph. While most attention models mainly re-weight or assign near-zero weights during training, this allows for explicitly and dynamically truncate the input sequence. Specifically:

Y = Softmax(Qψ_S(K)ᵀ_([:n]))ψ_S(V)_([:n])

where n is the Sortfut budget hyperparameter.

PaperSourceSee Code · lucidrains/sinkhorn-transformer

Papers archive 2025-07-28

1 shown of 1, 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

5 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
Document Classification1
Image Generation1
Language Modeling1
Language Modelling1
Natural Language Inference1

Usage over time archive 2025-07-28

Papers per year tagged with SortCut Sinkhorn Attention: 2020 to 2020, peak 1 1 0 2020: 1 paper 2020
Papers per year the archive tags with this method, by the paper's archive date (1 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

Attention Mechanisms

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