{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "paszke-et-al-pytorch-system-2019",
    "title": "Paszke and colleagues: PyTorch imperative deep-learning runtime",
    "summary": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components.",
    "boundary": "Year denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.",
    "area": "Machine-learning systems and compilers",
    "topic": "Computer science",
    "year": 2019,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "paszke-et-al-pytorch-system-2019",
    "target": {
      "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
      "version": 1,
      "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
    },
    "humans": [
      "Adam Paszke",
      "Sam Gross",
      "Francisco Massa",
      "Adam Lerer",
      "James Bradbury",
      "Gregory Chanan",
      "Trevor Killeen",
      "Zeming Lin",
      "Natalia Gimelshein",
      "Luca Antiga",
      "Alban Desmaison",
      "Andreas Köpf",
      "Edward Yang",
      "Zach DeVito",
      "Martin Raison",
      "Alykhan Tejani",
      "Sasank Chilamkurthy",
      "Benoit Steiner",
      "Lu Fang",
      "Junjie Bai",
      "Soumith Chintala"
    ],
    "authors_complete": true,
    "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.",
    "agents": [],
    "related": [],
    "subresults": [],
    "scientific_assessment": {
      "source_review": "SOURCE_CURATED",
      "scientific_review": "NOT_REVIEWED_BY_WRS",
      "formal_replay": "NOT_RUN",
      "provider_attestation": "NONE"
    },
    "sources": [
      {
        "id": "cur20261015-arxiv-1912-01703v1",
        "title": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
        "url": "https://arxiv.org/abs/1912.01703v1"
      }
    ],
    "points": "225000",
    "published_grade": {
      "id": "paszke-et-al-pytorch-system-2019",
      "entry_id": "paszke-et-al-pytorch-system-2019",
      "band": "A",
      "modifier": "+",
      "target": {
        "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
        "version": 1,
        "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
      },
      "what_changed": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components.",
      "scope": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components. Year denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.",
      "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:36e395e8-6408-48fd-ab55-ab6379273b1d",
        "version": 1,
        "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
      },
      "edition": "ratings-e039106f30dec177"
    }
  },
  "version": {
    "kind": "PUBLIC_RECORD_VIEW",
    "previous": null,
    "published_at": "2026-10-16T00:00:00Z",
    "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": "Adam Paszke",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Sam Gross",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Francisco Massa",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Adam Lerer",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "James Bradbury",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Gregory Chanan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Trevor Killeen",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zeming Lin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Natalia Gimelshein",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Luca Antiga",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Alban Desmaison",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Andreas Köpf",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Edward Yang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zach DeVito",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Martin Raison",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Alykhan Tejani",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Sasank Chilamkurthy",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Benoit Steiner",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Lu Fang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Junjie Bai",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Soumith Chintala",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
            "url": "https://arxiv.org/abs/1912.01703v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Year denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.",
        "domain": "Machine-learning systems and compilers"
      },
      "sources": [
        {
          "citation": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
          "url": "https://arxiv.org/abs/1912.01703v1"
        }
      ],
      "statement": "# Paszke and colleagues: PyTorch imperative deep-learning runtime\n\nCombined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components.\n\n## Scientific scope\n\nYear denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.\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\nAdam Paszke; Sam Gross; Francisco Massa; Adam Lerer; James Bradbury; Gregory Chanan; Trevor Killeen; Zeming Lin; Natalia Gimelshein; Luca Antiga; Alban Desmaison; Andreas Köpf; Edward Yang; Zach DeVito; Martin Raison; Alykhan Tejani; Sasank Chilamkurthy; Benoit Steiner; Lu Fang; Junjie Bai; Soumith Chintala\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://arxiv.org/abs/1912.01703v1)\n",
      "summary": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components.",
      "title": "Paszke and colleagues: PyTorch imperative deep-learning runtime"
    },
    "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
      "version": 1,
      "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:a74674f39e6e9b418bd556a7f4bbec84e2a20a0032a99708a824d7ef7792aa40"
  },
  "target": {
    "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
    "version": 1,
    "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "paszke-et-al-pytorch-system-2019",
    "entry_id": "paszke-et-al-pytorch-system-2019",
    "revision": 1,
    "title": "Paszke and colleagues: PyTorch imperative deep-learning runtime",
    "context": "Machine-learning systems and compilers",
    "track": "Human-designed research method or resource",
    "band": "A",
    "modifier": "+",
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components.",
    "scope": "Combined an imperative Python execution model with automatic differentiation and efficient accelerator-aware runtime components. Year denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.",
    "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Year denotes this system paper, not the first software release. The paper does not invent automatic differentiation or guarantee every eager program matches compiled performance.",
    "references": [
      {
        "title": "PyTorch: An Imperative Style, High-Performance Deep Learning Library",
        "url": "https://arxiv.org/abs/1912.01703v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      }
    ],
    "roles": [
      {
        "who": "Source-listed authors and technical contributors",
        "credit": "SCIENCE",
        "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities."
      }
    ],
    "target": {
      "record_id": "urn:uuid:36e395e8-6408-48fd-ab55-ab6379273b1d",
      "version": 1,
      "content_hash": "sha256:cc9f201cdf5d97bfd5e7051dc4e02478b2da17feed9ff5a9d1842e38a9756d42"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "225000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}