{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "bai-et-al-longbench-2023",
    "title": "LongBench: bilingual multi-task long-context evaluation",
    "summary": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models.",
    "boundary": "Combines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.",
    "area": "Long-context evaluation",
    "topic": "Artificial intelligence",
    "year": 2023,
    "lane": "HUMAN",
    "result": "BENCHMARK",
    "contribution_kind": "EVALUATION_RESOURCE_DESIGN",
    "package_id": "bai-et-al-longbench-2023",
    "target": {
      "record_id": "urn:uuid:ede86648-7a28-4304-ab31-4e1650278017",
      "version": 1,
      "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
    },
    "humans": [
      "Yushi Bai",
      "Xin Lv",
      "Jiajie Zhang",
      "Hongchang Lyu",
      "Jiankai Tang",
      "Zhidian Huang",
      "Zhengxiao Du",
      "Xiao Liu",
      "Aohan Zeng",
      "Lei Hou",
      "Yuxiao Dong",
      "Jie Tang",
      "Juanzi Li"
    ],
    "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-2308-14508v1",
        "title": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
        "url": "https://arxiv.org/abs/2308.14508v1"
      }
    ],
    "points": "125000",
    "published_grade": {
      "id": "bai-et-al-longbench-2023",
      "entry_id": "bai-et-al-longbench-2023",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:ede86648-7a28-4304-ab31-4e1650278017",
        "version": 1,
        "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
      },
      "what_changed": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models.",
      "scope": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models. Combines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.",
      "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:ede86648-7a28-4304-ab31-4e1650278017",
        "version": 1,
        "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
      },
      "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": "Yushi Bai",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xin Lv",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jiajie Zhang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Hongchang Lyu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jiankai Tang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zhidian Huang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zhengxiao Du",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xiao Liu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Aohan Zeng",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Lei Hou",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yuxiao Dong",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jie Tang",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Juanzi Li",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
            "url": "https://arxiv.org/abs/2308.14508v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Combines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.",
        "domain": "Long-context evaluation"
      },
      "sources": [
        {
          "citation": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
          "url": "https://arxiv.org/abs/2308.14508v1"
        }
      ],
      "statement": "# LongBench: bilingual multi-task long-context evaluation\n\nStandardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models.\n\n## Scientific scope\n\nCombines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.\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\nYushi Bai; Xin Lv; Jiajie Zhang; Hongchang Lyu; Jiankai Tang; Zhidian Huang; Zhengxiao Du; Xiao Liu; Aohan Zeng; Lei Hou; Yuxiao Dong; Jie Tang; Juanzi Li\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- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://arxiv.org/abs/2308.14508v1)\n",
      "summary": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models.",
      "title": "LongBench: bilingual multi-task long-context evaluation"
    },
    "record_id": "urn:uuid:ede86648-7a28-4304-ab31-4e1650278017",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:ede86648-7a28-4304-ab31-4e1650278017",
      "version": 1,
      "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:8620ac079e78aeb7c92763d2d29aec13c24fa37c40a554a2c48919d9bd5cdfd3"
  },
  "target": {
    "record_id": "urn:uuid:ede86648-7a28-4304-ab31-4e1650278017",
    "version": 1,
    "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "bai-et-al-longbench-2023",
    "entry_id": "bai-et-al-longbench-2023",
    "revision": 1,
    "title": "LongBench: bilingual multi-task long-context evaluation",
    "context": "Long-context evaluation",
    "track": "Evaluation resource design",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models.",
    "scope": "Standardized English and Chinese datasets across question answering, summarization, few-shot learning, synthetic tasks and code completion to compare long-context language models. Combines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.",
    "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. Combines existing and newly constructed tasks; constituent dataset authorship is not transferred to the suite authors. English lengths are words and Chinese lengths are characters in the original table, not uniform tokenizer counts. Task metrics and truncation policies must remain explicit.",
    "references": [
      {
        "title": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding",
        "url": "https://arxiv.org/abs/2308.14508v1",
        "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:ede86648-7a28-4304-ab31-4e1650278017",
      "version": 1,
      "content_hash": "sha256:4e9cac0e655c561a6b39f05075e2793fd141385b8856b7706966bd48a39f1748"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "125000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}