{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-11-01-007",
    "source_story_id": "tmp-story-strata-clinical-hypotheses",
    "edition_id": "mp-2026-11-01-morning-0114",
    "edition_url": "https://themachinepress.com/edition/2026-11-01",
    "position": 7,
    "story_type": "dispatch",
    "section": "benchmarks-evals",
    "editorial_classification": "editorial",
    "headline": "A Correct Verdict Can Still Use the Wrong Patients",
    "slug": "a-correct-verdict-can-still-use-the-wrong-patients",
    "dek": "STRATA grades plans, evidence sets, p-values and verdicts for 8,735 temporal hypotheses drawn from vaccine adverse-event timelines.",
    "summary": "STRATA grades plans, evidence sets, p-values and verdicts for 8,735 temporal hypotheses drawn from vaccine adverse-event timelines.",
    "body_text": "A fluent answer about whether one event follows another can hide a missing statistical test or the wrong evidence set. STRATA provides 8,735 temporal clinical hypotheses over vaccine adverse-event timelines, each with a gold plan, evidence set, p-value and verdict reproduced by a deterministic oracle. Across five model configurations, systems failed mainly when deciding which records counted, even when the final verdict looked correct. A staged articulate-retrieve-test workflow closed most of that gap and brought a 31-billion-parameter open model within 0.01 of frontier models under the same setup. The benchmark evaluates analysis procedure; it does not establish causal vaccine effects.",
    "why_it_matters": "STRATA grades plans, evidence sets, p-values and verdicts for 8,735 temporal hypotheses drawn from vaccine adverse-event timelines.",
    "limitations": [
      "The benchmark evaluates analysis procedure; it does not establish causal vaccine effects."
    ],
    "importance": 9,
    "canonical_url": "https://themachinepress.com/story/mp-2026-11-01-007/a-correct-verdict-can-still-use-the-wrong-patients",
    "json_url": "https://themachinepress.com/story/mp-2026-11-01-007.json",
    "first_published_at": "2026-11-01T09:00:00.000-05:00",
    "modified_at": "2026-11-01T09:00:00.000-05:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-11-01-007-001",
        "text": "STRATA grades plans, evidence sets, p-values and verdicts for 8,735 temporal hypotheses drawn from vaccine adverse-event timelines.",
        "source_ids": [
          "source-2026-11-01-007"
        ],
        "qualification": "The benchmark evaluates analysis procedure; it does not establish causal vaccine effects."
      }
    ],
    "source_ids": [
      "source-2026-11-01-007"
    ],
    "tags": [
      "clinical data",
      "statistical testing",
      "evidence",
      "benchmarks"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-11-01-007",
      "title": "arXiv preprint 2610.40619, original v1",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2610.40619v1",
      "canonical_url": "https://arxiv.org/abs/2610.40619v1",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-10-28T20:00:00.000-04:00",
      "accessed_at": "2026-11-01T08:20:00.000-05:00",
      "supports_claim_ids": [
        "claim-mp-2026-11-01-007-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "A Correct Verdict Can Still Use the Wrong Patients",
    "publisher": "The Machine Press",
    "published_at": "2026-11-01T09:00:00.000-05:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-11-01-007/a-correct-verdict-can-still-use-the-wrong-patients"
  }
}