{
  "schema_version": "2.0",
  "record_type": "dataset",
  "tier": "verified",
  "slug": "pubmedqa",
  "name": "PubMedQA",
  "summary": {
    "en": "License: permissive (MIT); Mdexa hosting: link-only. first-hand data-scope statement(s) allow commercial use and redistribution",
    "zh": "许可证声明：宽松（MIT）；本站处理：只链接原出处。针对数据的声明允许商用和再分发。"
  },
  "systems": [
    "modern-medicine"
  ],
  "domains": [
    "biomedical-literature",
    "clinical-medicine"
  ],
  "content_types": [
    "qa-dialogue",
    "annotated-corpus"
  ],
  "content_description": {
    "en": "to answer research questions with yes/no/maybe (e.g.: Do preoperative statins reduce atrial fibrillation after coronary artery bypass grafting?) using the corresponding abstracts.",
    "zh": "依据相应摘要，用是/否/可能来回答研究问题。"
  },
  "languages": [
    "en"
  ],
  "modalities": [
    "text"
  ],
  "size": {
    "raw": "100K<n<1M, 10K<n<100K, 1K<n<10K",
    "measures": []
  },
  "publisher": "qiaojin",
  "year": {
    "value": null,
    "raw": null
  },
  "license": {
    "legacy_status": "pending",
    "spdx": "MIT",
    "commercial_use": "yes",
    "derivatives": "yes",
    "redistribution": "yes",
    "rationale": {
      "en": "first-hand data-scope statement(s) allow commercial use and redistribution",
      "zh": "针对数据的声明允许商用和再分发"
    },
    "evidence": [
      {
        "location": "hf-card-metadata",
        "location_detail": {
          "en": "Statement found in snapshot s18 (api-hf)",
          "zh": "在快照 s18（api-hf）中找到的声明"
        },
        "excerpt": "\"mit\"",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA",
        "verified_on": "2026-11-12",
        "snapshot_id": "s18",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "readme",
        "location_detail": {
          "en": "Statement found in snapshot s16 (readme)",
          "zh": "在快照 s16（readme）中找到的声明"
        },
        "excerpt": "license:\n- mit",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA",
        "verified_on": "2026-11-12",
        "snapshot_id": "s16",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "dataset-page",
        "location_detail": {
          "en": "Statement found in snapshot s17 (page)",
          "zh": "在快照 s17（page）中找到的声明"
        },
        "excerpt": "License: mit",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA",
        "verified_on": "2026-11-12",
        "snapshot_id": "s17",
        "observed": "page",
        "captured_by": "pipeline"
      }
    ],
    "license_conditions": [
      "notice-required"
    ]
  },
  "license_status": "permissive",
  "hosting": "link-only",
  "hosting_reason": {
    "en": "License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Link only.",
    "zh": "许可证宽松：针对数据的声明允许商用和再分发。只链接原出处。"
  },
  "signoff": null,
  "risk_flags": [],
  "provenance": {
    "source_description": null,
    "upstream": []
  },
  "known_issues": [
    {
      "en": "The dataset card leaves the known limitations section as a placeholder rather than describing limitations.",
      "zh": "数据集卡片对已知局限性部分仅留占位符，未作说明。"
    }
  ],
  "preflight": {
    "status": "not-run"
  },
  "access": {
    "mdexa_download": null,
    "external_urls": [
      "https://huggingface.co/datasets/qiaojin/PubMedQA"
    ],
    "notes": null
  },
  "citation": {
    "text": null,
    "references": [
      "arXiv:1909.06146"
    ]
  },
  "verification_level": "verified",
  "last_verified": "2026-11-14",
  "tags": [
    {
      "facet": "task_format",
      "value": "choice-question",
      "basis": "author-stated",
      "excerpt": "task_ids:\n- multiple-choice-qa",
      "snapshot_id": "s16"
    },
    {
      "facet": "intended_use",
      "value": "evaluation",
      "basis": "author-stated",
      "excerpt": "500 questions in the `pqa_labeled` are used as the test set.",
      "snapshot_id": "s16"
    },
    {
      "facet": "benchmark_status",
      "value": "public-leaderboard",
      "basis": "author-stated",
      "excerpt": "The official leaderboard is available at: https://pubmedqa.github.io/.",
      "snapshot_id": "s16"
    },
    {
      "facet": "specialty",
      "value": "multi-specialty",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "excerpt": "to answer research questions with yes/no/maybe (e.g.: Do preoperative statins reduce atrial fibrillation after coronary artery bypass grafting?) using the corresponding abstracts.",
      "snapshot_id": "s16"
    },
    {
      "facet": "source_type",
      "value": "literature",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "excerpt": "using the corresponding abstracts",
      "snapshot_id": "s16"
    }
  ],
  "changelog": [
    {
      "date": "2026-11-14",
      "change": {
        "en": "Added via the automated pipeline and passed all checks (gates, verbatim quotes, rule engine, two independent extractions).",
        "zh": "经自动流水线收录并通过全部校验（门禁、逐字引用、判定引擎、两轮独立抽取）。"
      }
    }
  ],
  "access_recipe": {
    "verified_on": "2026-11-14",
    "platform": "huggingface",
    "repo_id": "qiaojin/PubMedQA",
    "source": {
      "api_url": "https://huggingface.co/api/datasets/qiaojin/PubMedQA",
      "fetched_on": "2026-11-14"
    },
    "access_state": "open",
    "files": [
      {
        "name": ".gitattributes",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA/resolve/9001f2853fb87cab8d220904e0de81ac6973b318/.gitattributes",
        "size_bytes": 1174,
        "checksum": null,
        "format": "gitattributes"
      },
      {
        "name": "README.md",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA/resolve/9001f2853fb87cab8d220904e0de81ac6973b318/README.md",
        "size_bytes": 5192,
        "checksum": null,
        "format": "md"
      },
      {
        "name": "pqa_artificial/train-00000-of-00001.parquet",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA/resolve/9001f2853fb87cab8d220904e0de81ac6973b318/pqa_artificial/train-00000-of-00001.parquet",
        "size_bytes": 233411194,
        "checksum": {
          "algorithm": "sha256",
          "value": "cdee2d3c01369f47d2ea7b89cd32ee64b142cc0debcb863bf39b3a81a3cece05"
        },
        "format": "parquet"
      },
      {
        "name": "pqa_labeled/train-00000-of-00001.parquet",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA/resolve/9001f2853fb87cab8d220904e0de81ac6973b318/pqa_labeled/train-00000-of-00001.parquet",
        "size_bytes": 1075513,
        "checksum": {
          "algorithm": "sha256",
          "value": "3d56bd1abc11884579ecc3aab9dd3cdfce8b7cf54715daeca93b8022b7be231c"
        },
        "format": "parquet"
      },
      {
        "name": "pqa_unlabeled/train-00000-of-00001.parquet",
        "url": "https://huggingface.co/datasets/qiaojin/PubMedQA/resolve/9001f2853fb87cab8d220904e0de81ac6973b318/pqa_unlabeled/train-00000-of-00001.parquet",
        "size_bytes": 66010017,
        "checksum": {
          "algorithm": "sha256",
          "value": "2f390ea5e6e3dc53b2736a52be2f594e05f5d8a599c3de754ad72f1068251e6e"
        },
        "format": "parquet"
      }
    ],
    "files_truncated": false,
    "commands": [
      {
        "command": "hf download qiaojin/PubMedQA --repo-type dataset",
        "evidence": {
          "excerpt": "You can use the hf download command from the terminal to directly download files from the Hub.",
          "doc": "hf-download",
          "url": "https://huggingface.co/docs/huggingface_hub/guides/download",
          "fetched_on": "2026-11-14"
        }
      }
    ],
    "registration": null,
    "version": {
      "commit": "9001f2853fb87cab8d220904e0de81ac6973b318",
      "last_modified": "2024-03-06T01:50:16.000Z"
    },
    "license_note": {
      "license_status": "permissive",
      "text": "Use according to the terms of the official source; this recipe does not say the data can be used freely."
    },
    "gaps": []
  },
  "_links": {
    "text": "在本库继续查找。本条目是 109 条已处理并核实的记录之一（数据集、数据资源与开源模型）。查找其他条目：调用 `https://mcp.mdexa.com/v1/datasets?q=<关键词>`，或连接 MCP 服务 `https://mcp.mdexa.com/mcp`；每分钟 20 次以内无需密钥。每行返回不到 1 KB；空结果即\"数据库未收录\"。核对一份清单：一次批量核对（免费账号）即返回报告。说明：/agents.md。相似条目：PMC-Patients-ReCDS (https://mdexa.com/zh/datasets/pmc-patients-recds/)；MedMCQA (https://mdexa.com/zh/datasets/medmcqa/)；CMB (Chinese Medical Benchmark) (https://mdexa.com/zh/datasets/cmb/)。",
    "records": 109,
    "search": {
      "template": "https://mcp.mdexa.com/v1/datasets?q={words}",
      "example": "https://mcp.mdexa.com/v1/datasets?q=%E7%BB%93%E5%90%88%E4%BA%B2%E5%92%8C%E5%8A%9B"
    },
    "mcp": "https://mcp.mdexa.com/mcp",
    "anonymous_calls_per_minute": 20,
    "guide": "https://mdexa.com/zh/agents.md",
    "faq": "https://mdexa.com/zh/developers/faq/",
    "similar": [
      {
        "name": "PMC-Patients-ReCDS",
        "url": "https://mdexa.com/zh/datasets/pmc-patients-recds/",
        "json": "https://mdexa.com/zh/datasets/pmc-patients-recds.json"
      },
      {
        "name": "MedMCQA",
        "url": "https://mdexa.com/zh/datasets/medmcqa/",
        "json": "https://mdexa.com/zh/datasets/medmcqa.json"
      },
      {
        "name": "CMB (Chinese Medical Benchmark)",
        "url": "https://mdexa.com/zh/datasets/cmb/",
        "json": "https://mdexa.com/zh/datasets/cmb.json"
      }
    ]
  }
}