Methods › General › Position Embeddings › Absolute Position Encodings

Absolute Position Encodings

13,942 papers tagged archive 2025-07-28

Introduced by Ashish Vaswani et al. in Attention Is All You Need

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

Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are added to the input embeddings at the bottoms of the encoder and decoder stacks. The positional encodings have the same dimension d_(model) as the embeddings, so that the two can be summed. In the original implementation, sine and cosine functions of different frequencies are used:

PE(pos, 2i) = sin(pos/10000^(2i/d_(model)))

PE(pos, 2i+1) = cos(pos/10000^(2i/d_(model)))

where pos is the position and i is the dimension. That is, each dimension of the positional encoding corresponds to a sinusoid. The wavelengths form a geometric progression from 2π to 10000 2̇π. This function was chosen because the authors hypothesized it would allow the model to easily learn to attend by relative positions, since for any fixed offset k, PEₚₒₛ₊ₖ can be represented as a linear function of PEₚₒₛ.

Image Source: D2L.ai

PaperSource

Papers archive 2025-07-28

30 shown of 13,942, 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 2,143 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 Modelling1,235
Decoder1,063
Language Modeling947
Translation758
Machine Translation718
Semantic Segmentation671
Object Detection545
Image Classification516
Question Answering511
object-detection494
Sentence454
Retrieval453
Segmentation441
Representation Learning422
image-classification414
Large Language Model409
Time Series373
Object348
Text Generation315
Transfer Learning293

Usage over time archive 2025-07-28

Papers per year tagged with Absolute Position Encodings: 2017 to 2025, peak 4,350 4,350 0 2017: 21 papers 2017 2018: 115 papers 2018 2019: 510 papers 2019 2020: 824 papers 2020 2021: 1470 papers 2021 2022: 1855 papers 2022 2023: 3216 papers 2023 2024: 4350 papers 2024 2025: 1581 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (13,942 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

Position Embeddings

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