{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "radford-et-al-gpt2-2019",
    "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling",
    "summary": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning.",
    "boundary": "The 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.",
    "area": "Language models",
    "topic": "Artificial intelligence",
    "year": 2019,
    "lane": "HUMAN",
    "result": "EMPIRICAL",
    "package_id": "radford-et-al-gpt2-2019",
    "target": {
      "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
      "version": 1,
      "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
    },
    "humans": [
      "Alec Radford",
      "Jeffrey Wu",
      "Rewon Child",
      "David Luan",
      "Dario Amodei",
      "Ilya Sutskever"
    ],
    "authors_complete": true,
    "human_role_note": "Complete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.",
    "agents": [],
    "related": [],
    "subresults": [],
    "scientific_assessment": {
      "source_review": "SOURCE_CURATED",
      "scientific_review": "NOT_REVIEWED_BY_WRS",
      "formal_replay": "NOT_RUN",
      "provider_attestation": "NONE"
    },
    "sources": [
      {
        "id": "openai-gpt2-paper-2019",
        "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
        "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
      },
      {
        "id": "openai-gpt2-page-2019",
        "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — company research page",
        "url": "https://openai.com/index/better-language-models/"
      }
    ],
    "points": "50000",
    "published_grade": {
      "id": "radford-et-al-gpt2-2019",
      "entry_id": "radford-et-al-gpt2-2019",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
        "version": 1,
        "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
      },
      "what_changed": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning.",
      "scope": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning. Scope limitation: The 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.",
      "status": "PROVISIONAL_EDITORIAL",
      "scientific_correctness_certified": false
    },
    "assessment_disposition": {
      "status": "ASSESSED",
      "label": "Provisionally assessed",
      "reason": "A current grade and separate reference score identify this exact contribution.",
      "target": {
        "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
        "version": 1,
        "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
      },
      "edition": "ratings-e039106f30dec177"
    }
  },
  "version": {
    "kind": "PUBLIC_RECORD_VIEW",
    "previous": null,
    "published_at": "2026-09-27T00:00:00Z",
    "publisher": {
      "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33",
      "role": "WRS_EDITORIAL_PUBLICATION"
    },
    "record": {
      "contributors": [
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Alec Radford",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jeffrey Wu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Rewon Child",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "David Luan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Dario Amodei",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ilya Sutskever",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
            "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
          }
        },
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-1",
          "kind": "source_reference",
          "source": {
            "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — company research page",
            "url": "https://openai.com/index/better-language-models/"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "The 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.",
        "domain": "Language models"
      },
      "sources": [
        {
          "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
          "url": "https://cdn.openai.com/better-language-models/language-models.pdf"
        },
        {
          "citation": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — company research page",
          "url": "https://openai.com/index/better-language-models/"
        }
      ],
      "statement": "# Radford and colleagues: GPT-2 and unsupervised multitask language modeling\n\nTrained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning.\n\n## Scientific scope\n\nThe 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nAlec Radford; Jeffrey Wu; Rewon Child; David Luan; Dario Amodei; Ilya Sutskever\n\nComplete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.\n\n\n\n## Original sources\n\n- [Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper](https://cdn.openai.com/better-language-models/language-models.pdf)\n- [Radford and colleagues: GPT-2 and unsupervised multitask language modeling — company research page](https://openai.com/index/better-language-models/)\n",
      "summary": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning.",
      "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling"
    },
    "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
      "version": 1,
      "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:2fed352319878deb0128823612c33ddf28d321b96e7526471f2c148be0bd0e54"
  },
  "target": {
    "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
    "version": 1,
    "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "radford-et-al-gpt2-2019",
    "entry_id": "radford-et-al-gpt2-2019",
    "revision": 1,
    "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling",
    "context": "Language models",
    "track": "Human-designed AI research method or resource",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-09-27",
    "evidence_cutoff": "2026-09-27",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning.",
    "scope": "Trained GPT-2 as a large autoregressive transformer on WebText and documented zero-shot transfer to several language tasks without task-specific fine-tuning. Scope limitation: The 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.",
    "evidence_summary": "The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. The 2019 paper reports task- and setup-specific zero-shot results; it is distinct from GPT-1 and GPT-3 and does not show general reasoning reliability.",
    "references": [
      {
        "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — original paper",
        "url": "https://cdn.openai.com/better-language-models/language-models.pdf",
        "locator": "Source-linked contribution and stated scientific limits; not an independent replication."
      },
      {
        "title": "Radford and colleagues: GPT-2 and unsupervised multitask language modeling — company research page",
        "url": "https://openai.com/index/better-language-models/",
        "locator": "Source-linked contribution and stated scientific limits; not an independent replication."
      }
    ],
    "roles": [
      {
        "who": "The named authors listed in the linked research record",
        "credit": "SCIENCE",
        "did": "Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured."
      }
    ],
    "target": {
      "record_id": "urn:uuid:b9b3af2d-05d1-433a-a5cd-05cee940ed9b",
      "version": 1,
      "content_hash": "sha256:5fb1f618e7d962557efeb44550234a4ae653b987a9d94e6023b9acc0b72bf147"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "50000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}