Methods › Natural Language Processing › Subword Segmentation › WordPiece

WordPiece

7,063 papers tagged archive 2025-07-28

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

WordPiece is a subword segmentation algorithm used in natural language processing. The vocabulary is initialized with individual characters in the language, then the most frequent combinations of symbols in the vocabulary are iteratively added to the vocabulary. The process is:

  1. Initialize the word unit inventory with all the characters in the text.
  2. Build a language model on the training data using the inventory from 1.
  3. Generate a new word unit by combining two units out of the current word inventory to increment the word unit inventory by one. Choose the new word unit out of all the possible ones that increases the likelihood on the training data the most when added to the model.
  4. Goto 2 until a predefined limit of word units is reached or the likelihood increase falls below a certain threshold.

Text: Source

Image: WordPiece as used in BERT

Source: Google's Neural Machine Translation System: Bridging the...

Papers archive 2025-07-28

30 shown of 7,063, 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 1,195 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
Retrieval1,242
RAG1,241
Language Modelling1,157
Retrieval-augmented Generation1,057
Language Modeling935
Question Answering758
Sentence747
Sentiment Analysis411
Text Classification393
text-classification347
Transfer Learning314
Classification263
Word Embeddings260
Named Entity Recognition248
Information Retrieval244
Named Entity Recognition (NER)237
named-entity-recognition237
Natural Language Understanding233
Large Language Model213
Articles193

Usage over time archive 2025-07-28

Papers per year tagged with WordPiece: 2016 to 2025, peak 1,399 1,399 0 2016: 1 paper 2016 2017: 1 paper 2017 2018: 6 papers 2018 2019: 578 papers 2019 2020: 1258 papers 2020 2021: 1343 papers 2021 2022: 847 papers 2022 2023: 860 papers 2023 2024: 1399 papers 2024 2025: 770 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (7,063 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

Subword Segmentation

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