{
  "schema": "metatron.intelligence.entity.v1",
  "id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
  "kind": "benchmark",
  "subtype": "live_human_preference_evaluation_policy",
  "external_identifiers": [
    {
      "scheme": "benchmark_definition_identifier",
      "value": "arena-leaderboard-policy-2026-09-01",
      "issuer_id": null,
      "canonical_uri": "https://arena.ai/blog/policy",
      "source": {
        "source_id": "BMRK-10",
        "url": "https://arena.ai/blog/policy"
      },
      "valid_from": null,
      "valid_to": null,
      "observed_at": "2026-09-03",
      "status": "observed"
    }
  ],
  "sources": [
    {
      "source_id": "BMRK-10",
      "url": "https://arena.ai/blog/policy",
      "title": "Arena Leaderboard Policy",
      "publisher": "Arena Team",
      "source_date": "2026-09-01",
      "effective_date": null,
      "retrieved_at": "2026-09-03",
      "content_sha256": "02e2d8fd39c3879019d576fa59aa7cb9aa222282f2a89c878c4a08eaaa0deb87",
      "capture_artifact": "rs8000:/opt/metatron/shared/agent-research/metatron-world-innovation-trend-atlas/raw/research/2026-09-03-wave83-benchmark-intelligence/source-captures/BMRK-10-arena-policy.html",
      "capture_method": "curl --location --fail with identified browser user agent; decoded response body preserved byte-for-byte",
      "final_url": "https://arena.ai/blog/policy",
      "http_status": 200,
      "media_type": "text/html",
      "byte_count": 199867
    }
  ],
  "statements": [
    {
      "statement_id": "STM-5B231A082AB7B5AD5F7F337730F8B15B",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "name",
      "value_or_object_id": "Arena Leaderboard Policy",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Page heading and `Last updated on 1 Sept 2026`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-B31A21AE4215CC3FD8093117F6CBAC22",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "benchmark_identifier",
      "value_or_object_id": "arena-leaderboard-policy-2026-09-01",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Page heading and `Last updated on 1 Sept 2026`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-92B1F0B5CCE8491BAFBD0F1E79091A78",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "version",
      "value_or_object_id": "last updated 2026-09-01",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Page heading and `Last updated on 1 Sept 2026`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-3C467F8E1F6D9C2C2E7BFBBC234C811B",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "publisher",
      "value_or_object_id": "Arena Team",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Page heading and `Last updated on 1 Sept 2026`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-2E06CCE892D3A0C41A5EF1A955200F52",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "task_scope",
      "value_or_object_id": "Live community evaluation of eligible foundation models through randomized battles and human preference votes across Arena surfaces.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Sections `Live and Community-Driven AI Model Evaluation` and `Listing models on the leaderboard`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-4E4DCBE0C044C0A2F2EE538B37779334",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "metric_contract",
      "value_or_object_id": "Leaderboard ratings are estimated from battle outcomes with a regression that reweights sampling; ranks and intervals depend on the current methodology and accumulated vote data.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Policy statements on sampling probability, reweighting and open ranking pipeline",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-73B1DD3131D8260F343BAF5BC07E754B",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "evaluation_conditions",
      "value_or_object_id": "Result identity includes Arena surface/category, policy and methodology revision, model endpoint identity and availability, battle/vote window, filters, sampling/reweighting and data-cleaning rules.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Transparency, listing, sampling, deprecation and data-sharing sections",
      "valid_from": null,
      "valid_to": null,
      "confidence": "source_reported",
      "status": "observed"
    },
    {
      "statement_id": "STM-4AC736043AF4A3A3B2B23D5C842A281B",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "comparability_boundary",
      "value_or_object_id": "The leaderboard is live and methodology changes are logged; a rank is not timeless and must retain its observation time, category, policy/method version, interval and vote window.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Transparency, listing, sampling, deprecation and data-sharing sections",
      "valid_from": null,
      "valid_to": null,
      "confidence": "editorial_synthesis",
      "status": "observed"
    },
    {
      "statement_id": "STM-31E242858190AD18C9D7E5B097E3E137",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "contamination_boundary",
      "value_or_object_id": "Live user prompts reduce dependence on one fixed public test set, but the policy does not establish absence of manipulation, selection effects, provider exposure or preference bias.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Transparency, listing, sampling, deprecation and data-sharing sections",
      "valid_from": null,
      "valid_to": null,
      "confidence": "editorial_synthesis",
      "status": "observed"
    },
    {
      "statement_id": "STM-9DE997FB644A1EDB118DE6A055EA1176",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "scope",
      "value_or_object_id": "Reviewed benchmark or evaluation-protocol identity; model results and leaderboard ranks are separate, unrepresented observations.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Page heading and `Last updated on 1 Sept 2026`",
      "valid_from": null,
      "valid_to": null,
      "confidence": "editorial_synthesis",
      "status": "observed"
    },
    {
      "statement_id": "STM-B8AC309507958E546D83D78B28F47A7E",
      "subject_id": "ENT-3a6be454-fedf-4237-8d71-f378e916b0a7",
      "predicate": "boundary",
      "value_or_object_id": "Human preference is not factual correctness, safety or universal capability. Model listing/deprecation and changing votes prevent a static rank from becoming an entity attribute.",
      "observed_at": "2026-09-03",
      "source_id": "BMRK-10",
      "source_locator": "Transparency, listing, sampling, deprecation and data-sharing sections",
      "valid_from": null,
      "valid_to": null,
      "confidence": "editorial_synthesis",
      "status": "observed"
    }
  ],
  "current_projection": {
    "name": {
      "value": "Arena Leaderboard Policy",
      "statement_ids": [
        "STM-5B231A082AB7B5AD5F7F337730F8B15B"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "benchmark_identifier": {
      "value": "arena-leaderboard-policy-2026-09-01",
      "statement_ids": [
        "STM-B31A21AE4215CC3FD8093117F6CBAC22"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "version": {
      "value": "last updated 2026-09-01",
      "statement_ids": [
        "STM-92B1F0B5CCE8491BAFBD0F1E79091A78"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "publisher": {
      "value": "Arena Team",
      "statement_ids": [
        "STM-3C467F8E1F6D9C2C2E7BFBBC234C811B"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "task_scope": {
      "value": "Live community evaluation of eligible foundation models through randomized battles and human preference votes across Arena surfaces.",
      "statement_ids": [
        "STM-2E06CCE892D3A0C41A5EF1A955200F52"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "metric_contract": {
      "value": "Leaderboard ratings are estimated from battle outcomes with a regression that reweights sampling; ranks and intervals depend on the current methodology and accumulated vote data.",
      "statement_ids": [
        "STM-4E4DCBE0C044C0A2F2EE538B37779334"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "evaluation_conditions": {
      "value": "Result identity includes Arena surface/category, policy and methodology revision, model endpoint identity and availability, battle/vote window, filters, sampling/reweighting and data-cleaning rules.",
      "statement_ids": [
        "STM-73B1DD3131D8260F343BAF5BC07E754B"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "comparability_boundary": {
      "value": "The leaderboard is live and methodology changes are logged; a rank is not timeless and must retain its observation time, category, policy/method version, interval and vote window.",
      "statement_ids": [
        "STM-4AC736043AF4A3A3B2B23D5C842A281B"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "contamination_boundary": {
      "value": "Live user prompts reduce dependence on one fixed public test set, but the policy does not establish absence of manipulation, selection effects, provider exposure or preference bias.",
      "statement_ids": [
        "STM-31E242858190AD18C9D7E5B097E3E137"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "scope": {
      "value": "Reviewed benchmark or evaluation-protocol identity; model results and leaderboard ranks are separate, unrepresented observations.",
      "statement_ids": [
        "STM-9DE997FB644A1EDB118DE6A055EA1176"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "boundary": {
      "value": "Human preference is not factual correctness, safety or universal capability. Model listing/deprecation and changing votes prevent a static rank from becoming an entity attribute.",
      "statement_ids": [
        "STM-B8AC309507958E546D83D78B28F47A7E"
      ],
      "selection_policy": "single_reviewed_statement"
    },
    "current_status": "unknown"
  },
  "claims": {
    "match_rule": "exact_source_url",
    "count": 0,
    "claim_ids": []
  },
  "relations": []
}