{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "zbontar-et-al-barlow-twins-2021",
    "title": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction",
    "summary": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions.",
    "boundary": "Redundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.",
    "area": "Self-supervised image representation learning",
    "topic": "Artificial intelligence",
    "year": 2021,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "zbontar-et-al-barlow-twins-2021",
    "target": {
      "record_id": "urn:uuid:84ec53ac-acb9-4232-aef5-375678958e60",
      "version": 1,
      "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
    },
    "humans": [
      "Jure Zbontar",
      "Li Jing",
      "Ishan Misra",
      "Yann LeCun",
      "Stéphane Deny"
    ],
    "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-2103-03230v1",
        "title": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
        "url": "https://arxiv.org/abs/2103.03230v1"
      },
      {
        "id": "cur20261015-support-zbontar-et-al-barlow-twins-2021",
        "title": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction — author/institution implementation",
        "url": "https://github.com/facebookresearch/barlowtwins"
      }
    ],
    "points": "115000",
    "published_grade": {
      "id": "zbontar-et-al-barlow-twins-2021",
      "entry_id": "zbontar-et-al-barlow-twins-2021",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:84ec53ac-acb9-4232-aef5-375678958e60",
        "version": 1,
        "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
      },
      "what_changed": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions.",
      "scope": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions. Redundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.",
      "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:84ec53ac-acb9-4232-aef5-375678958e60",
        "version": 1,
        "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
      },
      "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": "Jure Zbontar",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Li Jing",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ishan Misra",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yann LeCun",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Stéphane Deny",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
            "url": "https://arxiv.org/abs/2103.03230v1"
          }
        },
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-1",
          "kind": "source_reference",
          "source": {
            "citation": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction — author/institution implementation",
            "url": "https://github.com/facebookresearch/barlowtwins"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Redundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.",
        "domain": "Self-supervised image representation learning"
      },
      "sources": [
        {
          "citation": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
          "url": "https://arxiv.org/abs/2103.03230v1"
        },
        {
          "citation": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction — author/institution implementation",
          "url": "https://github.com/facebookresearch/barlowtwins"
        }
      ],
      "statement": "# Zbontar et al.: Barlow Twins cross-correlation redundancy reduction\n\nOptimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions.\n\n## Scientific scope\n\nRedundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.\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\nJure Zbontar; Li Jing; Ishan Misra; Yann LeCun; Stéphane Deny\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- [Barlow Twins: Self-Supervised Learning via Redundancy Reduction](https://arxiv.org/abs/2103.03230v1)\n- [Zbontar et al.: Barlow Twins cross-correlation redundancy reduction — author/institution implementation](https://github.com/facebookresearch/barlowtwins)\n",
      "summary": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions.",
      "title": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction"
    },
    "record_id": "urn:uuid:84ec53ac-acb9-4232-aef5-375678958e60",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:84ec53ac-acb9-4232-aef5-375678958e60",
      "version": 1,
      "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:93f7717fe4fc97623485ff4f1ecde2b756a37320772a22d35e7ed6d96bd6f0ae"
  },
  "target": {
    "record_id": "urn:uuid:84ec53ac-acb9-4232-aef5-375678958e60",
    "version": 1,
    "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "zbontar-et-al-barlow-twins-2021",
    "entry_id": "zbontar-et-al-barlow-twins-2021",
    "revision": 1,
    "title": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction",
    "context": "Self-supervised image representation learning",
    "track": "Human-designed research method or resource",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions.",
    "scope": "Optimized cross-view feature cross-correlation toward an identity matrix, combining invariance on matched dimensions with suppression of redundant dimensions. Redundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.",
    "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. Redundancy-reduction ideas predate this neural objective. No explicit negatives or asymmetric teacher are required by the recipe, but this is not proof of global optimization or universal transfer. Acquire machine-learning method, not a biological discovery.",
    "references": [
      {
        "title": "Barlow Twins: Self-Supervised Learning via Redundancy Reduction",
        "url": "https://arxiv.org/abs/2103.03230v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      },
      {
        "title": "Zbontar et al.: Barlow Twins cross-correlation redundancy reduction — author/institution implementation",
        "url": "https://github.com/facebookresearch/barlowtwins",
        "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:84ec53ac-acb9-4232-aef5-375678958e60",
      "version": 1,
      "content_hash": "sha256:65d24e2d322c738edc67080ac3a549106a12e7a1147a7b2582a3334e62455bb0"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "115000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}