{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "kuratov-et-al-babilong-2024",
    "title": "BABILong: reasoning facts embedded in long distractor text",
    "summary": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations.",
    "boundary": "Extends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.",
    "area": "Long-context reasoning evaluation",
    "topic": "Artificial intelligence",
    "year": 2024,
    "lane": "HUMAN",
    "result": "BENCHMARK",
    "contribution_kind": "EVALUATION_RESOURCE_DESIGN",
    "package_id": "kuratov-et-al-babilong-2024",
    "target": {
      "record_id": "urn:uuid:218965d4-c3e7-4daf-a94e-5c3c2333992d",
      "version": 1,
      "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
    },
    "humans": [
      "Yuri Kuratov",
      "Aydar Bulatov",
      "Petr Anokhin",
      "Ivan Rodkin",
      "Dmitry Sorokin",
      "Artyom Sorokin",
      "Mikhail Burtsev"
    ],
    "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-2406-10149v1",
        "title": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
        "url": "https://arxiv.org/abs/2406.10149v1"
      }
    ],
    "points": "65000",
    "published_grade": {
      "id": "kuratov-et-al-babilong-2024",
      "entry_id": "kuratov-et-al-babilong-2024",
      "band": "B",
      "modifier": "+",
      "target": {
        "record_id": "urn:uuid:218965d4-c3e7-4daf-a94e-5c3c2333992d",
        "version": 1,
        "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
      },
      "what_changed": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations.",
      "scope": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations. Extends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.",
      "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:218965d4-c3e7-4daf-a94e-5c3c2333992d",
        "version": 1,
        "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
      },
      "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": "Yuri Kuratov",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Aydar Bulatov",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Petr Anokhin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ivan Rodkin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Dmitry Sorokin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Artyom Sorokin",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Mikhail Burtsev",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
            "url": "https://arxiv.org/abs/2406.10149v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Extends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.",
        "domain": "Long-context reasoning evaluation"
      },
      "sources": [
        {
          "citation": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
          "url": "https://arxiv.org/abs/2406.10149v1"
        }
      ],
      "statement": "# BABILong: reasoning facts embedded in long distractor text\n\nEmbedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations.\n\n## Scientific scope\n\nExtends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.\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\nYuri Kuratov; Aydar Bulatov; Petr Anokhin; Ivan Rodkin; Dmitry Sorokin; Artyom Sorokin; Mikhail Burtsev\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- [BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack](https://arxiv.org/abs/2406.10149v1)\n",
      "summary": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations.",
      "title": "BABILong: reasoning facts embedded in long distractor text"
    },
    "record_id": "urn:uuid:218965d4-c3e7-4daf-a94e-5c3c2333992d",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:218965d4-c3e7-4daf-a94e-5c3c2333992d",
      "version": 1,
      "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:719c4c16d4c427015fe49ad1798d33e7ba2fa621cfa0d372fe698829b6ba46a1"
  },
  "target": {
    "record_id": "urn:uuid:218965d4-c3e7-4daf-a94e-5c3c2333992d",
    "version": 1,
    "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "kuratov-et-al-babilong-2024",
    "entry_id": "kuratov-et-al-babilong-2024",
    "revision": 1,
    "title": "BABILong: reasoning facts embedded in long distractor text",
    "context": "Long-context reasoning evaluation",
    "track": "Evaluation resource design",
    "band": "B",
    "modifier": "+",
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations.",
    "scope": "Embedded generated bAbI reasoning facts among ordered background book sentences, allowing controlled extension of context length while testing retrieval, fact chaining and other reasoning operations. Extends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.",
    "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. Extends inherited bAbI tasks and PG19 background data. The reported 11-million-token recurrent-memory experiment concerns in-domain single-fact answering, not all twenty tasks. Generating new samples reduces exact-example leakage but does not prove universal immunity to contamination or template learning.",
    "references": [
      {
        "title": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack",
        "url": "https://arxiv.org/abs/2406.10149v1",
        "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:218965d4-c3e7-4daf-a94e-5c3c2333992d",
      "version": 1,
      "content_hash": "sha256:d3aeb1abdc19fd727396730675d515b2679ecc3a300ebcbc000a3978b838971d"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "65000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}