{
  "kind": "mesh_review_f_score",
  "created_at": "2026-09-30T17:41:55+00:00",
  "snapshot": "snapshot:a87006a46b89c6f0fa3b22830c0cd07193548f256b0319552a3342b6a304df8a",
  "items": 80,
  "reviewer": [
    "codex"
  ],
  "reviewer_only": {
    "correct": 61,
    "incorrect": 16,
    "uncertain": 3,
    "precision": {
      "k": 61,
      "n": 77,
      "rate": 0.792,
      "ci95": [
        0.689,
        0.868
      ]
    }
  },
  "criteria": {
    "evidence": {
      "correct": 80
    },
    "ownership": {
      "correct": 77,
      "unclear": 3
    },
    "sdg_interpretation": {
      "correct": 62,
      "unclear": 2,
      "incorrect": 16
    },
    "claim_stage": {
      "correct": 77,
      "unclear": 3
    }
  },
  "by": {
    "navigation_eligible": {
      "true": {
        "correct": 51,
        "incorrect": 12,
        "uncertain": 1,
        "precision": {
          "k": 51,
          "n": 63,
          "rate": 0.81,
          "ci95": [
            0.696,
            0.888
          ]
        }
      },
      "false": {
        "correct": 10,
        "incorrect": 4,
        "uncertain": 2,
        "precision": {
          "k": 10,
          "n": 14,
          "rate": 0.714,
          "ci95": [
            0.454,
            0.883
          ]
        }
      }
    },
    "candidate_source": {
      "model_goal_nomination": {
        "correct": 3,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 3,
          "n": 3,
          "rate": 1.0,
          "ci95": [
            0.438,
            1
          ]
        }
      },
      "rebound_recorded_response": {
        "correct": 2,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 2,
          "n": 2,
          "rate": 1.0,
          "ci95": [
            0.342,
            1
          ]
        }
      },
      "retriever_authored_rubric": {
        "correct": 36,
        "incorrect": 9,
        "uncertain": 1,
        "precision": {
          "k": 36,
          "n": 45,
          "rate": 0.8,
          "ci95": [
            0.662,
            0.891
          ]
        }
      },
      "reassessment_hardened_rubric": {
        "correct": 14,
        "incorrect": 3,
        "uncertain": 0,
        "precision": {
          "k": 14,
          "n": 17,
          "rate": 0.824,
          "ci95": [
            0.59,
            0.938
          ]
        }
      },
      "model_target_nomination:rules": {
        "correct": 3,
        "incorrect": 1,
        "uncertain": 1,
        "precision": {
          "k": 3,
          "n": 4,
          "rate": 0.75,
          "ci95": [
            0.301,
            0.954
          ]
        }
      },
      "model_target_nomination:goal_nomination": {
        "correct": 3,
        "incorrect": 3,
        "uncertain": 1,
        "precision": {
          "k": 3,
          "n": 6,
          "rate": 0.5,
          "ci95": [
            0.188,
            0.812
          ]
        }
      }
    },
    "goal": {
      "1": {
        "correct": 2,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 2,
          "n": 2,
          "rate": 1.0,
          "ci95": [
            0.342,
            1
          ]
        }
      },
      "2": {
        "correct": 5,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 5,
          "n": 5,
          "rate": 1.0,
          "ci95": [
            0.566,
            1
          ]
        }
      },
      "3": {
        "correct": 6,
        "incorrect": 0,
        "uncertain": 1,
        "precision": {
          "k": 6,
          "n": 6,
          "rate": 1.0,
          "ci95": [
            0.61,
            1.0
          ]
        }
      },
      "4": {
        "correct": 4,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 4,
          "n": 4,
          "rate": 1.0,
          "ci95": [
            0.51,
            1
          ]
        }
      },
      "5": {
        "correct": 1,
        "incorrect": 1,
        "uncertain": 0,
        "precision": {
          "k": 1,
          "n": 2,
          "rate": 0.5,
          "ci95": [
            0.095,
            0.905
          ]
        }
      },
      "6": {
        "correct": 2,
        "incorrect": 0,
        "uncertain": 1,
        "precision": {
          "k": 2,
          "n": 2,
          "rate": 1.0,
          "ci95": [
            0.342,
            1
          ]
        }
      },
      "7": {
        "correct": 4,
        "incorrect": 2,
        "uncertain": 0,
        "precision": {
          "k": 4,
          "n": 6,
          "rate": 0.667,
          "ci95": [
            0.3,
            0.903
          ]
        }
      },
      "8": {
        "correct": 3,
        "incorrect": 1,
        "uncertain": 0,
        "precision": {
          "k": 3,
          "n": 4,
          "rate": 0.75,
          "ci95": [
            0.301,
            0.954
          ]
        }
      },
      "9": {
        "correct": 11,
        "incorrect": 1,
        "uncertain": 0,
        "precision": {
          "k": 11,
          "n": 12,
          "rate": 0.917,
          "ci95": [
            0.646,
            0.985
          ]
        }
      },
      "10": {
        "correct": 3,
        "incorrect": 1,
        "uncertain": 0,
        "precision": {
          "k": 3,
          "n": 4,
          "rate": 0.75,
          "ci95": [
            0.301,
            0.954
          ]
        }
      },
      "11": {
        "correct": 3,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 3,
          "n": 3,
          "rate": 1.0,
          "ci95": [
            0.438,
            1
          ]
        }
      },
      "12": {
        "correct": 4,
        "incorrect": 3,
        "uncertain": 0,
        "precision": {
          "k": 4,
          "n": 7,
          "rate": 0.571,
          "ci95": [
            0.25,
            0.842
          ]
        }
      },
      "13": {
        "correct": 6,
        "incorrect": 0,
        "uncertain": 0,
        "precision": {
          "k": 6,
          "n": 6,
          "rate": 1.0,
          "ci95": [
            0.61,
            1.0
          ]
        }
      },
      "14": {
        "correct": 2,
        "incorrect": 1,
        "uncertain": 0,
        "precision": {
          "k": 2,
          "n": 3,
          "rate": 0.667,
          "ci95": [
            0.208,
            0.939
          ]
        }
      },
      "15": {
        "correct": 1,
        "incorrect": 3,
        "uncertain": 0,
        "precision": {
          "k": 1,
          "n": 4,
          "rate": 0.25,
          "ci95": [
            0.046,
            0.699
          ]
        }
      },
      "16": {
        "correct": 2,
        "incorrect": 1,
        "uncertain": 1,
        "precision": {
          "k": 2,
          "n": 3,
          "rate": 0.667,
          "ci95": [
            0.208,
            0.939
          ]
        }
      },
      "17": {
        "correct": 2,
        "incorrect": 2,
        "uncertain": 0,
        "precision": {
          "k": 2,
          "n": 4,
          "rate": 0.5,
          "ci95": [
            0.15,
            0.85
          ]
        }
      }
    }
  },
  "adjudication": {
    "disputed": 19,
    "ruled": 19,
    "pending": [],
    "complete": true
  },
  "note": "Agreement with an AI reviewer (Codex) on a seeded goal-stratified sample of published supported links; adjudicated by the author where ruled. Not human-validated accuracy. Goal strata are sampled at different rates, so the pooled figure is unweighted.",
  "adjudicated": {
    "correct": 61,
    "incorrect": 14,
    "uncertain": 5,
    "precision": {
      "k": 61,
      "n": 75,
      "rate": 0.813,
      "ci95": [
        0.711,
        0.885
      ]
    },
    "by_navigation_eligible": {
      "true": {
        "correct": 51,
        "incorrect": 10,
        "uncertain": 3,
        "precision": {
          "k": 51,
          "n": 61,
          "rate": 0.836,
          "ci95": [
            0.724,
            0.908
          ]
        }
      },
      "false": {
        "correct": 10,
        "incorrect": 4,
        "uncertain": 2,
        "precision": {
          "k": 10,
          "n": 14,
          "rate": 0.714,
          "ci95": [
            0.454,
            0.883
          ]
        }
      }
    }
  }
}
