Neural Foundations · 2015

Neural Machine Translation of Rare Words with Subword Units (BPE)

Rico Sennrich, Barry Haddow, Alexandra Birch

Introduced byte pair encoding as a subword segmentation method for translation, letting a fixed vocabulary represent rare and unseen words as sequences of learned subword units and removing the out-of-vocabulary problem in open-vocabulary translation.

Editorial record

Plain-language summary

BPE starts from characters and iteratively merges the most frequent adjacent symbol pairs to build a vocabulary of subword units, so any word can be encoded as a sequence of known pieces. This let translation models handle rare words, compounds, and morphology without a huge word vocabulary or a separate back-off mechanism. Subword segmentation via BPE became standard tokenization for most later NMT and language models.

Knowledge graph

Relationships

Antecedents

  • Depends onEvidence: Direct

    Attention Is All You Need

    Subword (BPE) vocabulary for open-vocab modeling

    P-001 Sec 5.1

  • GeneralizesEvidence: Direct

    SentencePiece

    SentencePiece generalizes BPE to raw text

    P-120

Source record

Provenance

Record ID
A-013
Record created
2026-07-13
Last reviewed
2026-07-14
Record version
2

Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.