Methods › General › Attention Modules › Mixed Attention Block

Mixed Attention Block

6 papers tagged archive 2025-07-28

Introduced by Zi-Hang Jiang et al. in ConvBERT: Improving BERT with Span-based Dynamic Convolution

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

Mixed Attention Block is an attention module used in the ConvBERT architecture. It is a mixture of self-attention and span-based dynamic convolution (highlighted in pink). They share the same Query but use different Key to generate the attention map and convolution kernel respectively. The number of attention heads is reducing by directly projecting the input to a smaller embedding space to form a bottleneck structure for self-attention and span-based dynamic convolution. Dimensions of the input and output of some blocks are labeled on the left top corner to illustrate the overall framework, where d is the embedding size of the input and γ is the reduction ratio.

PaperSource

Papers archive 2025-07-28

6 shown of 6, 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

16 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
Automatic Speech Recognition1
Automatic Speech Recognition (ASR)1
Bias Detection1
Drug Discovery1
Language Modeling1
Language Modelling1
Misinformation1
Natural Language Understanding1
Object Detection1
Punctuation Restoration1
RGB Salient Object Detection1
Salient Object Detection1
Speech Recognition1
Transfer Learning1
object-detection1
speech-recognition1

Usage over time archive 2025-07-28

Papers per year tagged with Mixed Attention Block: 2020 to 2025, peak 2 2 0 2020: 1 paper 2020 2021: 0 papers 2021 2022: 0 papers 2022 2023: 2 papers 2023 2024: 1 paper 2024 2025: 2 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (6 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 Modules

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