{
  "apiVersion": "v1",
  "methodology": "https://sourcescore.org/methodology/",
  "canonical": "https://sourcescore.org/claims/aede848e23c8de8e/",
  "claim": {
    "vertical": "ai-ml",
    "subject": "Byte-Pair Encoding (BPE) for NMT",
    "predicate": "introduced_in_paper",
    "object": "Neural Machine Translation of Rare Words with Subword Units (Sennrich et al., 2015)",
    "confidence": 1,
    "sources": [
      {
        "url": "https://arxiv.org/abs/1508.07909",
        "title": "Neural Machine Translation of Rare Words with Subword Units",
        "publisher": "arXiv (Sennrich, Haddow, Birch)",
        "publishedDate": "2015-08-31",
        "accessedDate": "2026-05-16",
        "type": "preprint",
        "excerpt": "We discuss the suitability of different word segmentation techniques, including simple character n-gram models and a segmentation based on the byte pair encoding compression algorithm, and empirically show that subword models improve over a back-off dictionary baseline."
      },
      {
        "url": "https://aclanthology.org/P16-1162/",
        "title": "Neural Machine Translation of Rare Words with Subword Units (ACL 2016)",
        "publisher": "ACL Anthology",
        "publishedDate": "2016-08-07",
        "accessedDate": "2026-05-16",
        "type": "peer-reviewed"
      }
    ],
    "publishedAt": "2026-05-16T00:00:00Z",
    "lastVerified": "2026-05-16",
    "methodologyVersion": "veritas-v0.1",
    "tags": [
      "bpe",
      "tokenization",
      "subword",
      "foundational",
      "2015",
      "acl"
    ],
    "id": "aede848e23c8de8e",
    "statement": "Byte-Pair Encoding (BPE) for NMT introduced in paper: Neural Machine Translation of Rare Words with Subword Units (Sennrich et al., 2015)."
  },
  "signature": {
    "algorithm": "HMAC-SHA256",
    "signedBy": "did:web:sourcescore.org",
    "signedAt": "2026-05-16T00:00:00.000Z",
    "signature": "d33819d2484c7373b21a6c851fd49bafdff80c88adbd909694228dbcd558237e"
  },
  "citedAs": "Byte-Pair Encoding (BPE) for NMT introduced in paper: Neural Machine Translation of Rare Words with Subword Units (Sennrich et al., 2015). — SourceScore Claim aede848e23c8de8e (verified 2026-05-16, signed d33819d2…). https://sourcescore.org/claims/aede848e23c8de8e/"
}