{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
    "title": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining",
    "summary": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding.",
    "boundary": "December 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.",
    "area": "Multimodal document representation learning",
    "topic": "Artificial intelligence",
    "year": 2020,
    "lane": "HUMAN",
    "result": "EMPIRICAL",
    "package_id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
    "target": {
      "record_id": "urn:uuid:18745dea-4e43-4698-ad6d-f5b266ea8f27",
      "version": 1,
      "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
    },
    "humans": [
      "Yang Xu",
      "Yiheng Xu",
      "Tengchao Lv",
      "Lei Cui",
      "Furu Wei",
      "Guoxin Wang",
      "Yijuan Lu",
      "Dinei Florencio",
      "Cha Zhang",
      "Wanxiang Che",
      "Min Zhang",
      "Lidong Zhou"
    ],
    "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-2012-14740v1",
        "title": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
        "url": "https://arxiv.org/abs/2012.14740v1"
      },
      {
        "id": "cur20261015-support-xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
        "title": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining — official implementation",
        "url": "https://github.com/microsoft/unilm/tree/master/layoutlmv2"
      }
    ],
    "points": "75000",
    "published_grade": {
      "id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
      "entry_id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
      "band": "B",
      "modifier": "+",
      "target": {
        "record_id": "urn:uuid:18745dea-4e43-4698-ad6d-f5b266ea8f27",
        "version": 1,
        "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
      },
      "what_changed": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding.",
      "scope": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding. December 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.",
      "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:18745dea-4e43-4698-ad6d-f5b266ea8f27",
        "version": 1,
        "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
      },
      "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": "Yang Xu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yiheng Xu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Tengchao Lv",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Lei Cui",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Furu Wei",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Guoxin Wang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yijuan Lu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Dinei Florencio",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Cha Zhang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Wanxiang Che",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Min Zhang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Lidong Zhou",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
            "url": "https://arxiv.org/abs/2012.14740v1"
          }
        },
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-1",
          "kind": "source_reference",
          "source": {
            "citation": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining — official implementation",
            "url": "https://github.com/microsoft/unilm/tree/master/layoutlmv2"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "December 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.",
        "domain": "Multimodal document representation learning"
      },
      "sources": [
        {
          "citation": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
          "url": "https://arxiv.org/abs/2012.14740v1"
        },
        {
          "citation": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining — official implementation",
          "url": "https://github.com/microsoft/unilm/tree/master/layoutlmv2"
        }
      ],
      "statement": "# Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining\n\nAdded text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding.\n\n## Scientific scope\n\nDecember 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.\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\nYang Xu; Yiheng Xu; Tengchao Lv; Lei Cui; Furu Wei; Guoxin Wang; Yijuan Lu; Dinei Florencio; Cha Zhang; Wanxiang Che; Min Zhang; Lidong Zhou\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- [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://arxiv.org/abs/2012.14740v1)\n- [Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining — official implementation](https://github.com/microsoft/unilm/tree/master/layoutlmv2)\n",
      "summary": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding.",
      "title": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining"
    },
    "record_id": "urn:uuid:18745dea-4e43-4698-ad6d-f5b266ea8f27",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:18745dea-4e43-4698-ad6d-f5b266ea8f27",
      "version": 1,
      "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:4149c0559fa831d26d893755f9df8bb4e772a47eb92a799a3b8ba34868cb8322"
  },
  "target": {
    "record_id": "urn:uuid:18745dea-4e43-4698-ad6d-f5b266ea8f27",
    "version": 1,
    "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
    "entry_id": "xu-et-al-layoutlmv2-spatial-crossmodal-pretraining-2020",
    "revision": 1,
    "title": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining",
    "context": "Multimodal document representation learning",
    "track": "Human-designed research method or resource",
    "band": "B",
    "modifier": "+",
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding.",
    "scope": "Added text-image alignment and matching objectives with spatial-aware attention to jointly learn text, layout and image interactions for document understanding. December 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.",
    "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. December 2020 preprint precedes 2021 conference. Extends LayoutLM with specific cross-modal objectives/attention rather than inventing multimodal pretraining; benchmark gains are not production reliability certification.",
    "references": [
      {
        "title": "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding",
        "url": "https://arxiv.org/abs/2012.14740v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      },
      {
        "title": "Xu et al.: LayoutLMv2 spatial cross-modal alignment pretraining — official implementation",
        "url": "https://github.com/microsoft/unilm/tree/master/layoutlmv2",
        "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:18745dea-4e43-4698-ad6d-f5b266ea8f27",
      "version": 1,
      "content_hash": "sha256:5d1bf5d396fcfb7d81d8129d68586b2429e73c697587c9259f1fd5b8943160eb"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "75000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}