{
  "schema_version": "2.0",
  "record_type": "dataset",
  "tier": "verified",
  "slug": "medqa-2",
  "name": "MedQA (USMLE 4-option, US subset + English textbook corpus)",
  "summary": {
    "en": "License: permissive (MIT); Mdexa hosting: link-only. first-hand data-scope statement(s) allow commercial use and redistribution Risk flag(s): derived, exam-derived.",
    "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。来源风险标记：derived、exam-derived。"
  },
  "systems": [
    "modern-medicine"
  ],
  "domains": [
    "clinical-medicine"
  ],
  "content_types": [
    "benchmark-exam",
    "literature"
  ],
  "content_description": {
    "en": "Re-upload of the English USMLE 4-option question subset and the English textbook corpus (18 textbooks) from the original MedQA release, with the official train/validation/test split.",
    "zh": "原始 MedQA 数据集中英文 USMLE 四选一题目子集与英文教科书语料（18 本教材）的重新上传，含官方训练/验证/测试划分。"
  },
  "languages": [
    "en"
  ],
  "modalities": [
    "text"
  ],
  "size": {
    "raw": "10K<n<100K",
    "measures": []
  },
  "publisher": "awinml",
  "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 s01 (readme)",
          "zh": "在快照 s01（readme）中找到的声明"
        },
        "excerpt": "license: mit",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "readme",
        "location_detail": {
          "en": "Statement found in snapshot s01 (readme)",
          "zh": "在快照 s01（readme）中找到的声明"
        },
        "excerpt": "The original [jind11/MedQA](https://github.com/jind11/MedQA) repository is distributed under the [MIT License](https://github.com/jind11/MedQA/blob/master/LICENSE). This re-upload follows that license.",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "readme",
        "location_detail": {
          "en": "Statement found in snapshot s01 (readme)",
          "zh": "在快照 s01（readme）中找到的声明"
        },
        "excerpt": "Users should review the original repository and paper and ensure their intended use is compatible with any terms that apply to the underlying source materials.",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "readme",
        "location_detail": {
          "en": "Statement found in snapshot s01 (readme)",
          "zh": "在快照 s01（readme）中找到的声明"
        },
        "excerpt": "This dataset should **not** be used for clinical diagnosis, treatment recommendations, or as a substitute for licensed medical expertise.",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "dataset-page",
        "location_detail": {
          "en": "Statement found in snapshot s02 (page)",
          "zh": "在快照 s02（page）中找到的声明"
        },
        "excerpt": "This dataset should not be used for clinical diagnosis, treatment recommendations, or as a substitute for licensed medical expertise. Performance on multiple-choice exam questions does not reflect clinical safety.",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s02",
        "observed": "page",
        "captured_by": "pipeline"
      },
      {
        "location": "dataset-page",
        "location_detail": {
          "en": "Statement found in snapshot s02 (page)",
          "zh": "在快照 s02（page）中找到的声明"
        },
        "excerpt": "The original jind11/MedQA repository is distributed under the MIT License . This re-upload follows that license.",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/awinml/medqa",
        "verified_on": "2026-11-12",
        "snapshot_id": "s02",
        "observed": "page",
        "captured_by": "pipeline"
      }
    ],
    "license_conditions": [
      "notice-required"
    ],
    "notes": {
      "en": "risk flag(s) derived,exam-derived: never eligible or hosted",
      "zh": "risk flag(s) derived,exam-derived: never eligible or hosted"
    }
  },
  "license_status": "permissive",
  "hosting": "link-only",
  "hosting_reason": {
    "en": "License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): derived, exam-derived. Link only.",
    "zh": "许可证宽松：针对数据的声明允许商用和再分发。来源风险标记：derived、exam-derived。只链接原出处。"
  },
  "signoff": null,
  "risk_flags": [
    "derived",
    "exam-derived"
  ],
  "provenance": {
    "source_description": null,
    "upstream": [
      {
        "slug": null,
        "name": "jind11/MedQA",
        "relation": "derived-from"
      }
    ]
  },
  "known_issues": [
    {
      "en": "US-centric: contains only the English USMLE portion of MedQA.",
      "zh": "以美国为中心：仅包含 MedQA 的英文 USMLE 部分。"
    },
    {
      "en": "metamap_phrases are generated automatically and may be noisy or incomplete.",
      "zh": "metamap_phrases 由程序自动生成，可能含有噪声或不完整。"
    },
    {
      "en": "The textbook corpus is provided as raw text without chapter or section boundaries.",
      "zh": "教科书语料以原始文本提供，没有章节边界。"
    }
  ],
  "usage_notes": [
    {
      "kind": "disclaimer",
      "excerpt": "This dataset should not be used for clinical diagnosis, treatment recommendations, or as a substitute for licensed medical expertise.",
      "url": "https://huggingface.co/datasets/awinml/medqa",
      "snapshot_id": "s02"
    },
    {
      "kind": "disclaimer",
      "excerpt": "This dataset should not be used for clinical diagnosis, treatment recommendations, or as a substitute for licensed medical expertise.",
      "url": "https://huggingface.co/datasets/awinml/medqa",
      "snapshot_id": "s02"
    },
    {
      "kind": "disclaimer",
      "excerpt": "Performance on multiple-choice exam questions does not reflect clinical safety.",
      "url": "https://huggingface.co/datasets/awinml/medqa",
      "snapshot_id": "s01"
    }
  ],
  "preflight": {
    "status": "not-run"
  },
  "access": {
    "mdexa_download": null,
    "external_urls": [
      "https://huggingface.co/datasets/awinml/medqa"
    ],
    "notes": null
  },
  "citation": {
    "text": null,
    "references": [
      "arXiv:2009.13081",
      "10.3390/app11146421"
    ]
  },
  "verification_level": "verified",
  "last_verified": "2026-11-14",
  "tags": [
    {
      "facet": "task_format",
      "value": "choice-question",
      "basis": "author-stated",
      "excerpt": "given a clinical vignette and four answer options, predict the correct option.",
      "snapshot_id": "s01"
    },
    {
      "facet": "intended_use",
      "value": "evaluation",
      "basis": "author-stated",
      "excerpt": "evaluate domain-specific language models on USMLE-style clinical reasoning.",
      "snapshot_id": "s01"
    },
    {
      "facet": "intended_use",
      "value": "retrieval-kb",
      "basis": "author-stated",
      "excerpt": "use the textbook corpus for open-domain medical question answering with retrieval.",
      "snapshot_id": "s01"
    },
    {
      "facet": "specialty",
      "value": "multi-specialty",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "excerpt": "| Immunology_Janeway | Janeway's Immunobiology |\n| InternalMed_Harrison | Harrison's Principles of Internal Medicine |\n| Neurology_Adams | Adams and Victor's Principles of Neurology |\n| Obstentrics_Williams | Williams Obstetrics |",
      "snapshot_id": "s01"
    },
    {
      "facet": "source_type",
      "value": "textbook-reference",
      "basis": "author-stated",
      "excerpt": "Contains 18 medical textbooks used for retrieval-based open-domain QA.",
      "snapshot_id": "s01"
    },
    {
      "facet": "source_type",
      "value": "expert-written",
      "basis": "author-stated",
      "excerpt": "annotations_creators:\n- expert-generated",
      "snapshot_id": "s01"
    },
    {
      "facet": "official_split",
      "value": "full-split",
      "basis": "author-stated",
      "excerpt": "the official **train / validation / test** split",
      "snapshot_id": "s01"
    }
  ],
  "changelog": [
    {
      "date": "2026-11-13",
      "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）。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-14",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    },
    {
      "date": "2026-11-14",
      "change": {
        "en": "Re-adjudicated under the current rules; all machine checks still pass.",
        "zh": "按当前规则重新判定；机器检查仍全部通过。"
      }
    }
  ],
  "access_recipe": {
    "verified_on": "2026-11-14",
    "platform": "huggingface",
    "repo_id": "awinml/medqa",
    "source": {
      "api_url": "https://huggingface.co/api/datasets/awinml/medqa",
      "fetched_on": "2026-11-14"
    },
    "access_state": "open",
    "files": [
      {
        "name": ".gitattributes",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/.gitattributes",
        "size_bytes": 2574,
        "checksum": null,
        "format": "gitattributes"
      },
      {
        "name": "README.md",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/README.md",
        "size_bytes": 8981,
        "checksum": null,
        "format": "md"
      },
      {
        "name": "corpus/train.jsonl",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/corpus/train.jsonl",
        "size_bytes": 89886012,
        "checksum": {
          "algorithm": "sha256",
          "value": "d2821c564fb0fb514b594b1f25c9893d1113cb43c3ed5cc0e168982a8a529c8f"
        },
        "format": "jsonl"
      },
      {
        "name": "questions/test.jsonl",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/questions/test.jsonl",
        "size_bytes": 2075679,
        "checksum": null,
        "format": "jsonl"
      },
      {
        "name": "questions/train.jsonl",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/questions/train.jsonl",
        "size_bytes": 16211077,
        "checksum": {
          "algorithm": "sha256",
          "value": "3a65baf760f17c395058d6699315a7ce8aa767c50cb193fea87ceb94dd790324"
        },
        "format": "jsonl"
      },
      {
        "name": "questions/validation.jsonl",
        "url": "https://huggingface.co/datasets/awinml/medqa/resolve/afdb980b627374a7f10a0a11ed68455596c52742/questions/validation.jsonl",
        "size_bytes": 2024697,
        "checksum": null,
        "format": "jsonl"
      }
    ],
    "files_truncated": false,
    "commands": [
      {
        "command": "hf download awinml/medqa --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": "afdb980b627374a7f10a0a11ed68455596c52742",
      "last_modified": "2026-04-18T04:57:30.000Z"
    },
    "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": "Continue in this database. This record is one of 109 processed and verified records (datasets, data resources and open models). To find others, call `https://mcp.mdexa.com/v1/datasets?q=<words>` or connect the MCP server at `https://mcp.mdexa.com/mcp`; no key is needed for 20 calls per minute. Rows are under 1 KB; an empty result means \"not listed\". For a list of items, one batch check (free account) returns a report. Guide: /agents.md. Similar records: 医学考试题库 JSON 数据集 (https://mdexa.com/datasets/medical-exam-question-bank-json/); CMB (Chinese Medical Benchmark) (https://mdexa.com/datasets/cmb/); MedMCQA (https://mdexa.com/datasets/medmcqa/).",
    "records": 109,
    "search": {
      "template": "https://mcp.mdexa.com/v1/datasets?q={words}",
      "example": "https://mcp.mdexa.com/v1/datasets?q=binding%20affinity"
    },
    "mcp": "https://mcp.mdexa.com/mcp",
    "anonymous_calls_per_minute": 20,
    "guide": "https://mdexa.com/agents.md",
    "faq": "https://mdexa.com/developers/faq/",
    "similar": [
      {
        "name": "医学考试题库 JSON 数据集",
        "url": "https://mdexa.com/datasets/medical-exam-question-bank-json/",
        "json": "https://mdexa.com/datasets/medical-exam-question-bank-json.json"
      },
      {
        "name": "CMB (Chinese Medical Benchmark)",
        "url": "https://mdexa.com/datasets/cmb/",
        "json": "https://mdexa.com/datasets/cmb.json"
      },
      {
        "name": "MedMCQA",
        "url": "https://mdexa.com/datasets/medmcqa/",
        "json": "https://mdexa.com/datasets/medmcqa.json"
      }
    ]
  }
}