{
  "apiVersion": "v1",
  "methodology": "https://sourcescore.org/methodology/",
  "canonical": "https://sourcescore.org/claims/3e9122ba60a3fe99/",
  "claim": {
    "vertical": "ai-ml",
    "subject": "Grouped-Query Attention (GQA)",
    "predicate": "introduced_in_paper",
    "object": "GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (Ainslie et al., 2023)",
    "confidence": 0.92,
    "sources": [
      {
        "url": "https://arxiv.org/abs/2305.13245",
        "title": "GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints",
        "publisher": "arXiv (Ainslie, Lee-Thorp, de Jong, Zemlyanskiy, Lebrón, Sanghai — Google Research)",
        "publishedDate": "2023-05-22",
        "accessedDate": "2026-05-31",
        "type": "preprint",
        "excerpt": "introduce grouped-query attention (GQA), a generalization of multi-query attention which uses an intermediate (more than one, less than number of query heads) number of key-value heads."
      },
      {
        "url": "https://aclanthology.org/2023.emnlp-main.298/",
        "title": "GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (EMNLP 2023)",
        "publisher": "Association for Computational Linguistics",
        "publishedDate": "2023-12-06",
        "accessedDate": "2026-05-31",
        "type": "peer-reviewed"
      },
      {
        "url": "https://huggingface.co/papers/2305.13245",
        "title": "GQA (Hugging Face Papers)",
        "publisher": "Hugging Face",
        "accessedDate": "2026-05-31",
        "type": "docs"
      }
    ],
    "publishedAt": "2026-05-31T00:00:00Z",
    "lastVerified": "2026-05-31",
    "methodologyVersion": "veritas-v0.1",
    "tags": [
      "gqa",
      "grouped-query-attention",
      "attention",
      "transformer",
      "inference",
      "ainslie",
      "2023",
      "emnlp"
    ],
    "id": "3e9122ba60a3fe99",
    "statement": "Grouped-Query Attention (GQA) introduced in paper: GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (Ainslie et al., 2023)."
  },
  "signature": {
    "algorithm": "HMAC-SHA256",
    "signedBy": "did:web:sourcescore.org",
    "signedAt": "2026-05-31T00:00:00.000Z",
    "signature": "dd77613a23120dd71ce16a006d5de8f3900129239e635ddc410992e27efa454e"
  },
  "citedAs": "Grouped-Query Attention (GQA) introduced in paper: GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (Ainslie et al., 2023). — SourceScore Claim 3e9122ba60a3fe99 (verified 2026-05-31, signed dd77613a…). https://sourcescore.org/claims/3e9122ba60a3fe99/"
}