{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "wang-et-al-dynamic-resolution-mrope-2024",
    "title": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE",
    "summary": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model.",
    "boundary": "Original QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.",
    "area": "Vision-language model architectures",
    "topic": "Artificial intelligence",
    "year": 2024,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "wang-et-al-dynamic-resolution-mrope-2024",
    "target": {
      "record_id": "urn:uuid:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
      "version": 1,
      "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
    },
    "humans": [
      "Peng Wang",
      "Shuai Bai",
      "Sinan Tan",
      "Shijie Wang",
      "Zhihao Fan",
      "Jinze Bai",
      "Keqin Chen",
      "Xuejing Liu",
      "Jialin Wang",
      "Wenbin Ge",
      "Yang Fan",
      "Kai Dang",
      "Mengfei Du",
      "Xuancheng Ren",
      "Rui Men",
      "Dayiheng Liu",
      "Chang Zhou",
      "Jingren Zhou",
      "Junyang Lin"
    ],
    "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-2409-12191v1",
        "title": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
        "url": "https://arxiv.org/abs/2409.12191v1"
      },
      {
        "id": "cur20261015-support-wang-et-al-dynamic-resolution-mrope-2024",
        "title": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE — author/institutional project",
        "url": "https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct"
      }
    ],
    "points": "120000",
    "published_grade": {
      "id": "wang-et-al-dynamic-resolution-mrope-2024",
      "entry_id": "wang-et-al-dynamic-resolution-mrope-2024",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
        "version": 1,
        "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
      },
      "what_changed": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model.",
      "scope": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model. Original QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.",
      "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:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
        "version": 1,
        "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
      },
      "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": "Peng Wang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Shuai Bai",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Sinan Tan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Shijie Wang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zhihao Fan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jinze Bai",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Keqin Chen",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xuejing Liu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jialin Wang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Wenbin Ge",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yang Fan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Kai Dang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Mengfei Du",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xuancheng Ren",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Rui Men",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Dayiheng Liu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Chang Zhou",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jingren Zhou",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Junyang Lin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
            "url": "https://arxiv.org/abs/2409.12191v1"
          }
        },
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-1",
          "kind": "source_reference",
          "source": {
            "citation": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE — author/institutional project",
            "url": "https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Original QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.",
        "domain": "Vision-language model architectures"
      },
      "sources": [
        {
          "citation": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
          "url": "https://arxiv.org/abs/2409.12191v1"
        },
        {
          "citation": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE — author/institutional project",
          "url": "https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct"
        }
      ],
      "statement": "# Wang et al.: dynamic-resolution vision tokens and multimodal RoPE\n\nMapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model.\n\n## Scientific scope\n\nOriginal QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.\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\nPeng Wang; Shuai Bai; Sinan Tan; Shijie Wang; Zhihao Fan; Jinze Bai; Keqin Chen; Xuejing Liu; Jialin Wang; Wenbin Ge; Yang Fan; Kai Dang; Mengfei Du; Xuancheng Ren; Rui Men; Dayiheng Liu; Chang Zhou; Jingren Zhou; Junyang Lin\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- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://arxiv.org/abs/2409.12191v1)\n- [Wang et al.: dynamic-resolution vision tokens and multimodal RoPE — author/institutional project](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)\n",
      "summary": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model.",
      "title": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE"
    },
    "record_id": "urn:uuid:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
      "version": 1,
      "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:83a0f5739343103ac1fbd99f772f43990d72834223d2fd3340e661d68b02051a"
  },
  "target": {
    "record_id": "urn:uuid:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
    "version": 1,
    "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "wang-et-al-dynamic-resolution-mrope-2024",
    "entry_id": "wang-et-al-dynamic-resolution-mrope-2024",
    "revision": 1,
    "title": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE",
    "context": "Vision-language model architectures",
    "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": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model.",
    "scope": "Mapped varying image resolutions to variable visual-token counts and decomposed rotary position encoding across text, image and video axes within a unified visual-language model. Original QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.",
    "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. Original QwenVL is inherited; dynamic tokenization/M-RoPE is the scoped change. Native resolution is constrained by practical resizing/token budgets, not unlimited visual resolution or proven human-like perception. Current repository redirects to Qwen3VL, whose features are not backdated.",
    "references": [
      {
        "title": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution",
        "url": "https://arxiv.org/abs/2409.12191v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      },
      {
        "title": "Wang et al.: dynamic-resolution vision tokens and multimodal RoPE — author/institutional project",
        "url": "https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct",
        "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:eb5099bd-babb-4bbc-aceb-7b9f5d2246d1",
      "version": 1,
      "content_hash": "sha256:63d19f86fb1065b437c0843a6eff53f79a0ebf53cfc553847bf3a366c04cfc63"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "120000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}