{
  "version": "2026-10-30",
  "scope": "additional analyses for the author review items 5-9 and 13; saved data only, no model calls",
  "bootstrap": {
    "replicates": 10000,
    "seed": 20261030,
    "unit": "whole task"
  },
  "permutation": {
    "draws": 19999,
    "seed": 20261030,
    "rule": "paired swap of the two decision rules within a task"
  },
  "probabilityRanking": {
    "source": "https://eternalsearch.net/omega/experimental-support-20261022/report.json",
    "endpoint": "core: article correspondence by the majority of three cross-company judges",
    "baselineShare": 0.144,
    "bootstrap": {
      "replicates": 10000,
      "seed": 20261030,
      "unit": "task"
    },
    "models": [
      {
        "model": "claude-sonnet-5-5",
        "hypotheses": 250,
        "tasks": 50,
        "matches": 36,
        "auc": 0.7133307372793354,
        "aucCi": {
          "lo": 0.6246458758527844,
          "hi": 0.7933380153123427,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.45,
            "certain": 17,
            "tiedAtCut": 5,
            "freeSlots": 2,
            "expectedMatches": 7.4,
            "precision": 0.38947368421052636
          },
          {
            "k": 20,
            "cutoff": 0.45,
            "certain": 17,
            "tiedAtCut": 5,
            "freeSlots": 3,
            "expectedMatches": 7.6,
            "precision": 0.38
          },
          {
            "k": 21,
            "cutoff": 0.45,
            "certain": 17,
            "tiedAtCut": 5,
            "freeSlots": 4,
            "expectedMatches": 7.8,
            "precision": 0.37142857142857144
          },
          {
            "k": 22,
            "cutoff": 0.45,
            "certain": 17,
            "tiedAtCut": 5,
            "freeSlots": 5,
            "expectedMatches": 8.0,
            "precision": 0.36363636363636365
          }
        ],
        "uniqueProbabilities": 21
      },
      {
        "model": "claude-opus-5-5",
        "hypotheses": 215,
        "tasks": 43,
        "matches": 29,
        "auc": 0.7059695958472376,
        "aucCi": {
          "lo": 0.6019871588805076,
          "hi": 0.7938092148345388,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.62,
            "certain": 16,
            "tiedAtCut": 6,
            "freeSlots": 3,
            "expectedMatches": 7.0,
            "precision": 0.3684210526315789
          },
          {
            "k": 20,
            "cutoff": 0.62,
            "certain": 16,
            "tiedAtCut": 6,
            "freeSlots": 4,
            "expectedMatches": 7.333333333333333,
            "precision": 0.36666666666666664
          },
          {
            "k": 21,
            "cutoff": 0.62,
            "certain": 16,
            "tiedAtCut": 6,
            "freeSlots": 5,
            "expectedMatches": 7.666666666666667,
            "precision": 0.3650793650793651
          },
          {
            "k": 22,
            "cutoff": 0.62,
            "certain": 16,
            "tiedAtCut": 6,
            "freeSlots": 6,
            "expectedMatches": 8.0,
            "precision": 0.36363636363636365
          }
        ],
        "uniqueProbabilities": 40
      },
      {
        "model": "claude-haiku-5-5",
        "hypotheses": 250,
        "tasks": 50,
        "matches": 36,
        "auc": 0.6156542056074766,
        "aucCi": {
          "lo": 0.49968762769636804,
          "hi": 0.7168213286198262,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.4,
            "certain": 18,
            "tiedAtCut": 12,
            "freeSlots": 1,
            "expectedMatches": 5.25,
            "precision": 0.27631578947368424
          },
          {
            "k": 20,
            "cutoff": 0.4,
            "certain": 18,
            "tiedAtCut": 12,
            "freeSlots": 2,
            "expectedMatches": 5.5,
            "precision": 0.275
          },
          {
            "k": 21,
            "cutoff": 0.4,
            "certain": 18,
            "tiedAtCut": 12,
            "freeSlots": 3,
            "expectedMatches": 5.75,
            "precision": 0.27380952380952384
          },
          {
            "k": 22,
            "cutoff": 0.4,
            "certain": 18,
            "tiedAtCut": 12,
            "freeSlots": 4,
            "expectedMatches": 6.0,
            "precision": 0.2727272727272727
          }
        ],
        "uniqueProbabilities": 17
      },
      {
        "model": "claude-opus-4-8",
        "hypotheses": 240,
        "tasks": 48,
        "matches": 34,
        "auc": 0.6736150770988006,
        "aucCi": {
          "lo": 0.5635338781100159,
          "hi": 0.7662729922584822,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.58,
            "certain": 17,
            "tiedAtCut": 6,
            "freeSlots": 2,
            "expectedMatches": 6.333333333333333,
            "precision": 0.3333333333333333
          },
          {
            "k": 20,
            "cutoff": 0.58,
            "certain": 17,
            "tiedAtCut": 6,
            "freeSlots": 3,
            "expectedMatches": 6.5,
            "precision": 0.325
          },
          {
            "k": 21,
            "cutoff": 0.58,
            "certain": 17,
            "tiedAtCut": 6,
            "freeSlots": 4,
            "expectedMatches": 6.666666666666667,
            "precision": 0.3174603174603175
          },
          {
            "k": 22,
            "cutoff": 0.58,
            "certain": 17,
            "tiedAtCut": 6,
            "freeSlots": 5,
            "expectedMatches": 6.833333333333333,
            "precision": 0.3106060606060606
          }
        ],
        "uniqueProbabilities": 40
      },
      {
        "model": "gpt-6-astra",
        "hypotheses": 245,
        "tasks": 49,
        "matches": 35,
        "auc": 0.6824489795918367,
        "aucCi": {
          "lo": 0.5707748421560028,
          "hi": 0.7853967867062905,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.7,
            "certain": 17,
            "tiedAtCut": 2,
            "freeSlots": 2,
            "expectedMatches": 8.0,
            "precision": 0.42105263157894735
          },
          {
            "k": 20,
            "cutoff": 0.69,
            "certain": 19,
            "tiedAtCut": 1,
            "freeSlots": 1,
            "expectedMatches": 8.0,
            "precision": 0.4
          },
          {
            "k": 21,
            "cutoff": 0.68,
            "certain": 20,
            "tiedAtCut": 4,
            "freeSlots": 1,
            "expectedMatches": 8.25,
            "precision": 0.39285714285714285
          },
          {
            "k": 22,
            "cutoff": 0.68,
            "certain": 20,
            "tiedAtCut": 4,
            "freeSlots": 2,
            "expectedMatches": 8.5,
            "precision": 0.38636363636363635
          }
        ],
        "uniqueProbabilities": 60
      },
      {
        "model": "gemini-4-argon-high",
        "hypotheses": 250,
        "tasks": 50,
        "matches": 36,
        "auc": 0.7398753894080997,
        "aucCi": {
          "lo": 0.6439648427328004,
          "hi": 0.8202498801662997,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.78,
            "certain": 11,
            "tiedAtCut": 9,
            "freeSlots": 8,
            "expectedMatches": 6.777777777777778,
            "precision": 0.3567251461988304
          },
          {
            "k": 20,
            "cutoff": 0.78,
            "certain": 11,
            "tiedAtCut": 9,
            "freeSlots": 9,
            "expectedMatches": 7.0,
            "precision": 0.35
          },
          {
            "k": 21,
            "cutoff": 0.76,
            "certain": 20,
            "tiedAtCut": 2,
            "freeSlots": 1,
            "expectedMatches": 7.5,
            "precision": 0.35714285714285715
          },
          {
            "k": 22,
            "cutoff": 0.76,
            "certain": 20,
            "tiedAtCut": 2,
            "freeSlots": 2,
            "expectedMatches": 8.0,
            "precision": 0.36363636363636365
          }
        ],
        "uniqueProbabilities": 44
      },
      {
        "model": "grok-4.5",
        "hypotheses": 225,
        "tasks": 45,
        "matches": 31,
        "auc": 0.7140006651147323,
        "aucCi": {
          "lo": 0.6045127484157792,
          "hi": 0.8078059185841844,
          "valid": 10000
        },
        "topK": [
          {
            "k": 19,
            "cutoff": 0.6,
            "certain": 17,
            "tiedAtCut": 2,
            "freeSlots": 2,
            "expectedMatches": 7.0,
            "precision": 0.3684210526315789
          },
          {
            "k": 20,
            "cutoff": 0.59,
            "certain": 19,
            "tiedAtCut": 1,
            "freeSlots": 1,
            "expectedMatches": 7.0,
            "precision": 0.35
          },
          {
            "k": 21,
            "cutoff": 0.58,
            "certain": 20,
            "tiedAtCut": 1,
            "freeSlots": 1,
            "expectedMatches": 7.0,
            "precision": 0.3333333333333333
          },
          {
            "k": 22,
            "cutoff": 0.57,
            "certain": 21,
            "tiedAtCut": 1,
            "freeSlots": 1,
            "expectedMatches": 7.0,
            "precision": 0.3181818181818182
          }
        ],
        "uniqueProbabilities": 53
      }
    ]
  },
  "groupAgainstBestSingle": [
    {
      "label": "Grok 4.5",
      "group": {
        "selected": 19,
        "matches": 14,
        "known": 40,
        "ppv": 0.7368421052631579,
        "sens": 0.35
      },
      "singles": [
        {
          "model": "claude-sonnet-5-5",
          "selected": 46,
          "matches": 21,
          "known": 40,
          "ppv": 0.45652173913043476,
          "sens": 0.525
        },
        {
          "model": "gpt-6-astra",
          "selected": 53,
          "matches": 25,
          "known": 40,
          "ppv": 0.4716981132075472,
          "sens": 0.625
        },
        {
          "model": "gemini-4-argon-high",
          "selected": 52,
          "matches": 23,
          "known": 40,
          "ppv": 0.4423076923076923,
          "sens": 0.575
        },
        {
          "model": "grok-4.5",
          "selected": 110,
          "matches": 31,
          "known": 40,
          "ppv": 0.2818181818181818,
          "sens": 0.775
        }
      ],
      "best": {
        "model": "gpt-6-astra",
        "selected": 53,
        "matches": 25,
        "known": 40,
        "ppv": 0.4716981132075472,
        "sens": 0.625
      },
      "equalLength": {
        "listSize": 19,
        "expectedMatchesInRandomSublist": 8.962264150943396,
        "rule": "a uniform random sublist of the best model's yes list has the full list's match share in expectation, so the comparison at equal length is the comparison of shares"
      },
      "difference": 0.2651439920556107,
      "differenceCi": {
        "lo": 0.08169389978213512,
        "hi": 0.459093228200371,
        "valid": 10000
      },
      "permutation": {
        "difference": 0.2651439920556107,
        "p": 0.0071,
        "draws": 19999,
        "method": "random paired swaps",
        "pHolm": 0.0213,
        "family": "four sets of generated hypotheses"
      },
      "workload": {
        "first": 240,
        "survivors": 53,
        "additionalModels": 3,
        "second": 159,
        "total": 399,
        "fullPanel": 960,
        "saved": 561
      }
    },
    {
      "label": "Gemini 4 Argon",
      "group": {
        "selected": 19,
        "matches": 9,
        "known": 33,
        "ppv": 0.47368421052631576,
        "sens": 0.2727272727272727
      },
      "singles": [
        {
          "model": "claude-sonnet-5-5",
          "selected": 45,
          "matches": 12,
          "known": 33,
          "ppv": 0.26666666666666666,
          "sens": 0.36363636363636365
        },
        {
          "model": "gpt-6-astra",
          "selected": 51,
          "matches": 14,
          "known": 33,
          "ppv": 0.27450980392156865,
          "sens": 0.42424242424242425
        },
        {
          "model": "gemini-4-argon-high",
          "selected": 53,
          "matches": 18,
          "known": 33,
          "ppv": 0.33962264150943394,
          "sens": 0.5454545454545454
        },
        {
          "model": "grok-4.5",
          "selected": 105,
          "matches": 20,
          "known": 33,
          "ppv": 0.19047619047619047,
          "sens": 0.6060606060606061
        }
      ],
      "best": {
        "model": "gemini-4-argon-high",
        "selected": 53,
        "matches": 18,
        "known": 33,
        "ppv": 0.33962264150943394,
        "sens": 0.5454545454545454
      },
      "equalLength": {
        "listSize": 19,
        "expectedMatchesInRandomSublist": 6.452830188679245,
        "rule": "a uniform random sublist of the best model's yes list has the full list's match share in expectation, so the comparison at equal length is the comparison of shares"
      },
      "difference": 0.13406156901688182,
      "differenceCi": {
        "lo": -0.04297907074719386,
        "hi": 0.3148148148148148,
        "valid": 10000
      },
      "permutation": {
        "difference": 0.13406156901688182,
        "p": 0.08335,
        "draws": 19999,
        "method": "random paired swaps",
        "pHolm": 0.08335,
        "family": "four sets of generated hypotheses"
      },
      "workload": {
        "first": 250,
        "survivors": 51,
        "additionalModels": 3,
        "second": 153,
        "total": 403,
        "fullPanel": 1000,
        "saved": 597
      }
    },
    {
      "label": "Claude Opus 5.5",
      "group": {
        "selected": 22,
        "matches": 14,
        "known": 45,
        "ppv": 0.6363636363636364,
        "sens": 0.3111111111111111
      },
      "singles": [
        {
          "model": "claude-sonnet-5-5",
          "selected": 50,
          "matches": 19,
          "known": 45,
          "ppv": 0.38,
          "sens": 0.4222222222222222
        },
        {
          "model": "gpt-6-astra",
          "selected": 49,
          "matches": 21,
          "known": 45,
          "ppv": 0.42857142857142855,
          "sens": 0.4666666666666667
        },
        {
          "model": "gemini-4-argon-high",
          "selected": 56,
          "matches": 22,
          "known": 45,
          "ppv": 0.39285714285714285,
          "sens": 0.4888888888888889
        },
        {
          "model": "grok-4.5",
          "selected": 110,
          "matches": 26,
          "known": 45,
          "ppv": 0.23636363636363636,
          "sens": 0.5777777777777777
        }
      ],
      "best": {
        "model": "gpt-6-astra",
        "selected": 49,
        "matches": 21,
        "known": 45,
        "ppv": 0.42857142857142855,
        "sens": 0.4666666666666667
      },
      "equalLength": {
        "listSize": 22,
        "expectedMatchesInRandomSublist": 9.428571428571429,
        "rule": "a uniform random sublist of the best model's yes list has the full list's match share in expectation, so the comparison at equal length is the comparison of shares"
      },
      "difference": 0.2077922077922078,
      "differenceCi": {
        "lo": 0.06818181818181818,
        "hi": 0.3561709401709399,
        "valid": 10000
      },
      "permutation": {
        "difference": 0.2077922077922078,
        "p": 0.01075,
        "draws": 19999,
        "method": "random paired swaps",
        "pHolm": 0.0215,
        "family": "four sets of generated hypotheses"
      },
      "workload": {
        "first": 250,
        "survivors": 49,
        "additionalModels": 3,
        "second": 147,
        "total": 397,
        "fullPanel": 1000,
        "saved": 603
      }
    },
    {
      "label": "Claude Opus 5.5, повторная генерация",
      "group": {
        "selected": 17,
        "matches": 10,
        "known": 43,
        "ppv": 0.5882352941176471,
        "sens": 0.23255813953488372
      },
      "singles": [
        {
          "model": "claude-sonnet-5-5",
          "selected": 43,
          "matches": 14,
          "known": 43,
          "ppv": 0.32558139534883723,
          "sens": 0.32558139534883723
        },
        {
          "model": "gpt-6-astra",
          "selected": 49,
          "matches": 16,
          "known": 43,
          "ppv": 0.32653061224489793,
          "sens": 0.37209302325581395
        },
        {
          "model": "gemini-4-argon-high",
          "selected": 61,
          "matches": 21,
          "known": 43,
          "ppv": 0.3442622950819672,
          "sens": 0.4883720930232558
        },
        {
          "model": "grok-4.5",
          "selected": 113,
          "matches": 28,
          "known": 43,
          "ppv": 0.24778761061946902,
          "sens": 0.6511627906976745
        }
      ],
      "best": {
        "model": "gemini-4-argon-high",
        "selected": 61,
        "matches": 21,
        "known": 43,
        "ppv": 0.3442622950819672,
        "sens": 0.4883720930232558
      },
      "equalLength": {
        "listSize": 17,
        "expectedMatchesInRandomSublist": 5.852459016393443,
        "rule": "a uniform random sublist of the best model's yes list has the full list's match share in expectation, so the comparison at equal length is the comparison of shares"
      },
      "difference": 0.24397299903567987,
      "differenceCi": {
        "lo": 0.06039191312645979,
        "hi": 0.4473746867167919,
        "valid": 10000
      },
      "permutation": {
        "difference": 0.24397299903567987,
        "p": 0.00045,
        "draws": 19999,
        "method": "random paired swaps",
        "pHolm": 0.0018,
        "family": "four sets of generated hypotheses"
      },
      "workload": {
        "first": 245,
        "survivors": 49,
        "additionalModels": 3,
        "second": 147,
        "total": 392,
        "fullPanel": 980,
        "saved": 588
      }
    }
  ],
  "withoutOwnCompanyPredictor": [
    {
      "label": "Grok 4.5",
      "generatorCompany": "xAI",
      "dropped": [
        "grok-4.5"
      ],
      "members": [
        "claude-sonnet-5-5",
        "gpt-6-astra",
        "gemini-4-argon-high"
      ],
      "full": {
        "selected": 19,
        "matches": 14,
        "known": 40,
        "ppv": 0.7368421052631579,
        "sens": 0.35,
        "ppvCi": {
          "lo": 0.5263157894736842,
          "hi": 0.9285714285714286,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.25,
          "hi": 0.46511627906976744,
          "valid": 10000
        },
        "tasks": 48
      },
      "reduced": {
        "selected": 22,
        "matches": 15,
        "known": 40,
        "ppv": 0.6818181818181818,
        "sens": 0.375,
        "ppvCi": {
          "lo": 0.47619047619047616,
          "hi": 0.8666666666666667,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.2702702702702703,
          "hi": 0.5,
          "valid": 10000
        },
        "tasks": 48
      },
      "firstStageUnchanged": true
    },
    {
      "label": "Gemini 4 Argon",
      "generatorCompany": "Google",
      "dropped": [
        "gemini-4-argon-high"
      ],
      "members": [
        "claude-sonnet-5-5",
        "gpt-6-astra",
        "grok-4.5"
      ],
      "full": {
        "selected": 19,
        "matches": 9,
        "known": 33,
        "ppv": 0.47368421052631576,
        "sens": 0.2727272727272727,
        "ppvCi": {
          "lo": 0.25,
          "hi": 0.7058823529411765,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.13513513513513514,
          "hi": 0.425,
          "valid": 10000
        },
        "tasks": 50
      },
      "reduced": {
        "selected": 20,
        "matches": 9,
        "known": 33,
        "ppv": 0.45,
        "sens": 0.2727272727272727,
        "ppvCi": {
          "lo": 0.23076923076923078,
          "hi": 0.6666666666666666,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.13513513513513514,
          "hi": 0.425,
          "valid": 10000
        },
        "tasks": 50
      },
      "firstStageUnchanged": true
    },
    {
      "label": "Claude Opus 5.5",
      "generatorCompany": "Anthropic",
      "dropped": [
        "claude-sonnet-5-5"
      ],
      "members": [
        "gpt-6-astra",
        "gemini-4-argon-high",
        "grok-4.5"
      ],
      "full": {
        "selected": 22,
        "matches": 14,
        "known": 45,
        "ppv": 0.6363636363636364,
        "sens": 0.3111111111111111,
        "ppvCi": {
          "lo": 0.4375,
          "hi": 0.8214285714285714,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.18864976415094345,
          "hi": 0.4423076923076923,
          "valid": 10000
        },
        "tasks": 50
      },
      "reduced": {
        "selected": 28,
        "matches": 15,
        "known": 45,
        "ppv": 0.5357142857142857,
        "sens": 0.3333333333333333,
        "ppvCi": {
          "lo": 0.35,
          "hi": 0.7083333333333334,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.20754716981132076,
          "hi": 0.46875,
          "valid": 10000
        },
        "tasks": 50
      },
      "firstStageUnchanged": true
    },
    {
      "label": "Claude Opus 5.5, повторная генерация",
      "generatorCompany": "Anthropic",
      "dropped": [
        "claude-sonnet-5-5"
      ],
      "members": [
        "gpt-6-astra",
        "gemini-4-argon-high",
        "grok-4.5"
      ],
      "full": {
        "selected": 17,
        "matches": 10,
        "known": 43,
        "ppv": 0.5882352941176471,
        "sens": 0.23255813953488372,
        "ppvCi": {
          "lo": 0.3333333333333333,
          "hi": 0.8235294117647058,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.125,
          "hi": 0.34782608695652173,
          "valid": 10000
        },
        "tasks": 49
      },
      "reduced": {
        "selected": 28,
        "matches": 13,
        "known": 43,
        "ppv": 0.4642857142857143,
        "sens": 0.3023255813953488,
        "ppvCi": {
          "lo": 0.2608695652173913,
          "hi": 0.6562723214285711,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.16666666666666666,
          "hi": 0.4444940476190469,
          "valid": 10000
        },
        "tasks": 49
      },
      "firstStageUnchanged": true
    }
  ],
  "strictEnrichment": [
    {
      "label": "Grok 4.5",
      "core": {
        "selected": {
          "selected": 19,
          "matches": 14,
          "known": 40,
          "ppv": 0.7368421052631579,
          "sens": 0.35,
          "ppvCi": {
            "lo": 0.5263157894736842,
            "hi": 0.9285714285714286,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.25,
            "hi": 0.46511627906976744,
            "valid": 10000
          },
          "tasks": 48
        },
        "baselineShare": 0.16666666666666666,
        "baseline": {
          "matches": 40,
          "hypotheses": 240
        },
        "enrichment": 4.421052631578948,
        "enrichmentCi": {
          "lo": 3.2297405918889295,
          "hi": 6.346153846153846,
          "valid": 10000
        }
      },
      "strict": {
        "selected": {
          "selected": 19,
          "matches": 5,
          "known": 7,
          "ppv": 0.2631578947368421,
          "sens": 0.7142857142857143,
          "ppvCi": {
            "lo": 0.07142857142857142,
            "hi": 0.47368421052631576,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.3333333333333333,
            "hi": 1.0,
            "valid": 9989
          },
          "tasks": 48
        },
        "baselineShare": 0.029166666666666667,
        "baseline": {
          "matches": 7,
          "hypotheses": 240
        },
        "enrichment": 9.022556390977442,
        "enrichmentCi": {
          "lo": 4.2105263157894735,
          "hi": 16.0,
          "valid": 9989
        }
      }
    },
    {
      "label": "Gemini 4 Argon",
      "core": {
        "selected": {
          "selected": 19,
          "matches": 9,
          "known": 33,
          "ppv": 0.47368421052631576,
          "sens": 0.2727272727272727,
          "ppvCi": {
            "lo": 0.25,
            "hi": 0.7058823529411765,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.13513513513513514,
            "hi": 0.425,
            "valid": 10000
          },
          "tasks": 50
        },
        "baselineShare": 0.132,
        "baseline": {
          "matches": 33,
          "hypotheses": 250
        },
        "enrichment": 3.588516746411483,
        "enrichmentCi": {
          "lo": 2.0833333333333335,
          "hi": 5.347593582887701,
          "valid": 10000
        }
      },
      "strict": {
        "selected": {
          "selected": 19,
          "matches": 6,
          "known": 9,
          "ppv": 0.3157894736842105,
          "sens": 0.6666666666666666,
          "ppvCi": {
            "lo": 0.1111111111111111,
            "hi": 0.5384615384615384,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.36363636363636365,
            "hi": 1.0,
            "valid": 9999
          },
          "tasks": 50
        },
        "baselineShare": 0.036,
        "baseline": {
          "matches": 9,
          "hypotheses": 250
        },
        "enrichment": 8.771929824561404,
        "enrichmentCi": {
          "lo": 5.177514792899408,
          "hi": 14.705882352941178,
          "valid": 9999
        }
      }
    },
    {
      "label": "Claude Opus 5.5",
      "core": {
        "selected": {
          "selected": 22,
          "matches": 14,
          "known": 45,
          "ppv": 0.6363636363636364,
          "sens": 0.3111111111111111,
          "ppvCi": {
            "lo": 0.4375,
            "hi": 0.8214285714285714,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.18864976415094345,
            "hi": 0.4423076923076923,
            "valid": 10000
          },
          "tasks": 50
        },
        "baselineShare": 0.18,
        "baseline": {
          "matches": 45,
          "hypotheses": 250
        },
        "enrichment": 3.5353535353535355,
        "enrichmentCi": {
          "lo": 2.462121212121212,
          "hi": 5.10204081632653,
          "valid": 10000
        }
      },
      "strict": {
        "selected": {
          "selected": 22,
          "matches": 2,
          "known": 5,
          "ppv": 0.09090909090909091,
          "sens": 0.4,
          "ppvCi": {
            "lo": 0.0,
            "hi": 0.20833333333333334,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.0,
            "hi": 1.0,
            "valid": 9950
          },
          "tasks": 50
        },
        "baselineShare": 0.02,
        "baseline": {
          "matches": 5,
          "hypotheses": 250
        },
        "enrichment": 4.545454545454546,
        "enrichmentCi": {
          "lo": 0.0,
          "hi": 11.029411764705882,
          "valid": 9950
        }
      }
    },
    {
      "label": "Claude Opus 5.5, повторная генерация",
      "core": {
        "selected": {
          "selected": 17,
          "matches": 10,
          "known": 43,
          "ppv": 0.5882352941176471,
          "sens": 0.23255813953488372,
          "ppvCi": {
            "lo": 0.3333333333333333,
            "hi": 0.8235294117647058,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.125,
            "hi": 0.34782608695652173,
            "valid": 10000
          },
          "tasks": 49
        },
        "baselineShare": 0.17551020408163265,
        "baseline": {
          "matches": 43,
          "hypotheses": 245
        },
        "enrichment": 3.351573187414501,
        "enrichmentCi": {
          "lo": 2.1416083916083917,
          "hi": 4.921875,
          "valid": 10000
        }
      },
      "strict": {
        "selected": {
          "selected": 17,
          "matches": 2,
          "known": 3,
          "ppv": 0.11764705882352941,
          "sens": 0.6666666666666666,
          "ppvCi": {
            "lo": 0.0,
            "hi": 0.29411764705882354,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.0,
            "hi": 1.0,
            "valid": 9532
          },
          "tasks": 49
        },
        "baselineShare": 0.012244897959183673,
        "baseline": {
          "matches": 3,
          "hypotheses": 245
        },
        "enrichment": 9.607843137254902,
        "enrichmentCi": {
          "lo": 0.0,
          "hi": 18.846153846153847,
          "valid": 9532
        }
      }
    }
  ],
  "balancedSet": {
    "claims": 100,
    "control": {
      "rule": "deterministic lexical rule over the English claim text: a direction verb and the valence of the measured readout; experimental labels take no part",
      "classified": 61,
      "unclassified": 39,
      "confirmedClassified": 38,
      "refutedClassified": 23,
      "accuracyOnClassified": 0.6557377049180327,
      "favourableAmongConfirmed": 35,
      "favourableAmongRefuted": 18,
      "reversedDirectionNegatives": 49,
      "negatives": 50,
      "agreementWithHeuristic": [
        {
          "model": "claude-sonnet-5-5",
          "claims": 61,
          "same": 35,
          "share": 0.5737704918032787
        },
        {
          "model": "gpt-6-astra",
          "claims": 61,
          "same": 36,
          "share": 0.5901639344262295
        },
        {
          "model": "gemini-4-argon-high",
          "claims": 61,
          "same": 35,
          "share": 0.5737704918032787
        },
        {
          "model": "grok-4.5",
          "claims": 61,
          "same": 31,
          "share": 0.5081967213114754
        },
        {
          "model": "claude-opus-5-5",
          "claims": 61,
          "same": 38,
          "share": 0.6229508196721312
        },
        {
          "model": "coin",
          "claims": 61,
          "same": 39,
          "share": 0.639344262295082
        }
      ]
    },
    "pairs": [
      {
        "model": "claude-sonnet-5-5",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 26,
        "acceptedReversedRejectedTrue": 2,
        "acceptedBoth": 13,
        "rejectedBoth": 9,
        "share": 0.52
      },
      {
        "model": "gpt-6-astra",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 23,
        "acceptedReversedRejectedTrue": 2,
        "acceptedBoth": 18,
        "rejectedBoth": 7,
        "share": 0.46
      },
      {
        "model": "gemini-4-argon-high",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 31,
        "acceptedReversedRejectedTrue": 3,
        "acceptedBoth": 6,
        "rejectedBoth": 10,
        "share": 0.62
      },
      {
        "model": "grok-4.5",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 30,
        "acceptedReversedRejectedTrue": 5,
        "acceptedBoth": 6,
        "rejectedBoth": 9,
        "share": 0.6
      },
      {
        "model": "claude-opus-5-5",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 24,
        "acceptedReversedRejectedTrue": 1,
        "acceptedBoth": 18,
        "rejectedBoth": 7,
        "share": 0.48
      },
      {
        "model": "coin",
        "pairs": 50,
        "acceptedTrueRejectedReversed": 12,
        "acceptedReversedRejectedTrue": 10,
        "acceptedBoth": 15,
        "rejectedBoth": 13,
        "share": 0.24
      }
    ],
    "transfer": {
      "unanimousClaims": 26,
      "unanimousConfirmed": 23,
      "ppvOnBalancedSet": 0.8846153846153846,
      "sensitivity": 0.46,
      "specificity": 0.94,
      "formula": "PPV = Se × π / [Se × π + (1 − Sp) × (1 − π)]",
      "atGeneratedPrevalence": {
        "самая низкая доля соответствий при одном запросе": {
          "prevalence": 0.132,
          "ppv": 0.5382978723404254
        },
        "самая высокая доля соответствий при одном запросе": {
          "prevalence": 0.18,
          "ppv": 0.627272727272727
        }
      }
    },
    "screening": {
      "sources": 50,
      "commonBoundary": "2026-06-30",
      "earliestPublicDate": "2026-07-01",
      "latestPublicDate": "2026-10-05",
      "preprintRecheckRecords": 50,
      "preprintSearchSources": [
        "pubmed_update_of",
        "openalex_location",
        "crossref_relation",
        "europepmc_authors",
        "europepmc_title",
        "crossref_search",
        "openalex_search"
      ],
      "candidatesExcludedForScreening": 15,
      "evidence": "https://eternalsearch.net/omega/balanced-confirmation-20261022/reconstruction.zip: benchmark/preprint-recheck/*.json and benchmark/exclusions.json"
    },
    "accuracyByLabelling": [
      {
        "model": "claude-sonnet-5-5",
        "labelled": {
          "claims": 61,
          "correct": 42,
          "accuracy": 0.6885245901639344
        },
        "unlabelled": {
          "claims": 39,
          "correct": 32,
          "accuracy": 0.8205128205128205
        }
      },
      {
        "model": "gpt-6-astra",
        "labelled": {
          "claims": 61,
          "correct": 47,
          "accuracy": 0.7704918032786885
        },
        "unlabelled": {
          "claims": 39,
          "correct": 24,
          "accuracy": 0.6153846153846154
        }
      },
      {
        "model": "gemini-4-argon-high",
        "labelled": {
          "claims": 61,
          "correct": 46,
          "accuracy": 0.7540983606557377
        },
        "unlabelled": {
          "claims": 39,
          "correct": 32,
          "accuracy": 0.8205128205128205
        }
      },
      {
        "model": "grok-4.5",
        "labelled": {
          "claims": 61,
          "correct": 46,
          "accuracy": 0.7540983606557377
        },
        "unlabelled": {
          "claims": 39,
          "correct": 29,
          "accuracy": 0.7435897435897436
        }
      },
      {
        "model": "claude-opus-5-5",
        "labelled": {
          "claims": 61,
          "correct": 43,
          "accuracy": 0.7049180327868853
        },
        "unlabelled": {
          "claims": 39,
          "correct": 30,
          "accuracy": 0.7692307692307693
        }
      },
      {
        "model": "coin",
        "labelled": {
          "claims": 61,
          "correct": 32,
          "accuracy": 0.5245901639344263
        },
        "unlabelled": {
          "claims": 39,
          "correct": 20,
          "accuracy": 0.5128205128205128
        }
      }
    ],
    "pairsLabelledByRule": {
      "rule": {
        "pairs": 20,
        "acceptedTrueRejectedReversed": 5,
        "acceptedReversedRejectedTrue": 2,
        "acceptedBoth": 13,
        "rejectedBoth": 0
      },
      "models": [
        {
          "model": "claude-sonnet-5-5",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 9,
          "acceptedReversedRejectedTrue": 2,
          "acceptedBoth": 5,
          "rejectedBoth": 4
        },
        {
          "model": "gpt-6-astra",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 11,
          "acceptedReversedRejectedTrue": 0,
          "acceptedBoth": 5,
          "rejectedBoth": 4
        },
        {
          "model": "gemini-4-argon-high",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 11,
          "acceptedReversedRejectedTrue": 1,
          "acceptedBoth": 3,
          "rejectedBoth": 5
        },
        {
          "model": "grok-4.5",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 13,
          "acceptedReversedRejectedTrue": 1,
          "acceptedBoth": 2,
          "rejectedBoth": 4
        },
        {
          "model": "claude-opus-5-5",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 8,
          "acceptedReversedRejectedTrue": 1,
          "acceptedBoth": 7,
          "rejectedBoth": 4
        },
        {
          "model": "coin",
          "pairs": 20,
          "acceptedTrueRejectedReversed": 5,
          "acceptedReversedRejectedTrue": 3,
          "acceptedBoth": 6,
          "rejectedBoth": 6
        }
      ]
    }
  },
  "keyElements": {
    "rule": "an element counts as named when at least two thirds of its content tokens occur in the hypothesis text; lexical check, not a semantic judgement",
    "sets": [
      {
        "label": "Grok 4.5",
        "selected": 19,
        "meanNamedShareMatched": 0.11428571428571428,
        "meanNamedShareUnmatched": 0.0,
        "atLeastTwoElementsMatched": 0,
        "atLeastTwoElementsUnmatched": 0,
        "rows": [
          {
            "task": "R42605193",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42630044",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42647119",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42647635",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42665751",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42689491",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42696595",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42698274",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42717261",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42723546",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42805954",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42810392",
            "id": "H3",
            "articleMatch": 1,
            "elements": 4,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42842666",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42844664",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42858183",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42669067",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42811065",
            "id": "H1",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42806003",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42806572",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          }
        ]
      },
      {
        "label": "Gemini 4 Argon",
        "selected": 19,
        "meanNamedShareMatched": 0.3222222222222222,
        "meanNamedShareUnmatched": 0.1,
        "atLeastTwoElementsMatched": 4,
        "atLeastTwoElementsUnmatched": 2,
        "rows": [
          {
            "task": "R42605193",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42608398",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42627902",
            "id": "H5",
            "articleMatch": 0,
            "elements": 4,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42647119",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42647635",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42663085",
            "id": "H1",
            "articleMatch": 0,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42664348",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42665751",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 4,
            "share": 0.8
          },
          {
            "task": "R42689491",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42723546",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42736449",
            "id": "H4",
            "articleMatch": 1,
            "elements": 4,
            "named": 1,
            "share": 0.25
          },
          {
            "task": "R42775695",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42805954",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42810392",
            "id": "H4",
            "articleMatch": 1,
            "elements": 4,
            "named": 1,
            "share": 0.25
          },
          {
            "task": "R42826184",
            "id": "H1",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42842666",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42853938",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42858183",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42585340",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          }
        ]
      },
      {
        "label": "Claude Opus 5.5",
        "selected": 22,
        "meanNamedShareMatched": 0.13571428571428573,
        "meanNamedShareUnmatched": 0.05,
        "atLeastTwoElementsMatched": 2,
        "atLeastTwoElementsUnmatched": 0,
        "rows": [
          {
            "task": "R42630044",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42664348",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42665751",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 3,
            "share": 0.6
          },
          {
            "task": "R42689491",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42717261",
            "id": "H1",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42717261",
            "id": "H2",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42717261",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42723546",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42725837",
            "id": "H2",
            "articleMatch": 1,
            "elements": 4,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42736449",
            "id": "H2",
            "articleMatch": 1,
            "elements": 4,
            "named": 2,
            "share": 0.5
          },
          {
            "task": "R42749700",
            "id": "H2",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42775698",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42786310",
            "id": "H5",
            "articleMatch": 0,
            "elements": 4,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42841227",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42842666",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42844664",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42853938",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42723144",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42808804",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42669067",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42806003",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42748238",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          }
        ]
      },
      {
        "label": "Claude Opus 5.5, повторная генерация",
        "selected": 17,
        "meanNamedShareMatched": 0.24500000000000002,
        "meanNamedShareUnmatched": 0.1142857142857143,
        "atLeastTwoElementsMatched": 2,
        "atLeastTwoElementsUnmatched": 1,
        "rows": [
          {
            "task": "R42601407",
            "id": "H4",
            "articleMatch": 0,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42605193",
            "id": "H1",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42608398",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42630044",
            "id": "H2",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42647635",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 4,
            "share": 0.8
          },
          {
            "task": "R42717261",
            "id": "H3",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42723546",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 2,
            "share": 0.4
          },
          {
            "task": "R42749700",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42775698",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42786310",
            "id": "H4",
            "articleMatch": 1,
            "elements": 4,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42821663",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42822442",
            "id": "H4",
            "articleMatch": 1,
            "elements": 4,
            "named": 1,
            "share": 0.25
          },
          {
            "task": "R42858183",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 0,
            "share": 0.0
          },
          {
            "task": "R42723144",
            "id": "H4",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42585340",
            "id": "H3",
            "articleMatch": 0,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42669067",
            "id": "H5",
            "articleMatch": 1,
            "elements": 5,
            "named": 1,
            "share": 0.2
          },
          {
            "task": "R42806003",
            "id": "H5",
            "articleMatch": 0,
            "elements": 5,
            "named": 1,
            "share": 0.2
          }
        ]
      }
    ],
    "pooled": {
      "matched": 37,
      "unmatched": 23,
      "meanNamedShareMatched": 0.172972972972973,
      "meanNamedShareUnmatched": 0.0608695652173913,
      "atLeastTwoMatched": 6,
      "atLeastTwoUnmatched": 2
    }
  },
  "secondReview": {
    "astraSameSet": {
      "source": "https://eternalsearch.net/omega/direct-confirmation-grok-20261022/report.json (voting rows with all four short-question forecasts)",
      "hypotheses": 245,
      "tasks": 49,
      "matches": 36,
      "baselineShare": 0.1469387755102041,
      "stages": {
        "afterAstraYes": {
          "selected": 74,
          "matches": 23,
          "known": 36,
          "ppv": 0.3108108108108108,
          "sens": 0.6388888888888888
        },
        "twoStage": {
          "selected": 22,
          "matches": 11,
          "known": 36,
          "ppv": 0.5,
          "sens": 0.3055555555555556,
          "ppvCi": {
            "lo": 0.3,
            "hi": 0.7037581699346398,
            "valid": 10000
          },
          "sensCi": {
            "lo": 0.18181818181818182,
            "hi": 0.4375,
            "valid": 10000
          },
          "tasks": 49
        }
      },
      "twoStageEnrichment": 3.4027777777777777,
      "twoStageEnrichmentCi": {
        "lo": 2.2615167243367935,
        "hi": 4.954044117647059,
        "valid": 10000
      },
      "strict": {
        "twoStage": {
          "selected": 22,
          "matches": 3,
          "known": 4,
          "ppv": 0.13636363636363635,
          "sens": 0.75
        },
        "baseline": {
          "selected": 245,
          "matches": 4,
          "known": 4,
          "ppv": 0.0163265306122449,
          "sens": 1.0
        }
      },
      "coin": {
        "selected": 114,
        "matches": 17,
        "known": 36,
        "ppv": 0.14912280701754385,
        "sens": 0.4722222222222222,
        "ppvCi": {
          "lo": 0.09090909090909091,
          "hi": 0.21238938053097345,
          "valid": 10000
        },
        "sensCi": {
          "lo": 0.3333333333333333,
          "hi": 0.6296296296296297,
          "valid": 10000
        },
        "tasks": 49
      },
      "listSize": 22,
      "ranking": [
        {
          "model": "claude-sonnet-5-5",
          "withProbability": 245,
          "expectedMatches": 8.0,
          "precision": 0.36363636363636365,
          "enrichment": 2.4747474747474745,
          "enrichmentCi": {
            "lo": 1.2949260042283297,
            "hi": 4.000045372050816,
            "valid": 10000
          },
          "difference": 0.13636363636363635,
          "differenceCi": {
            "lo": -0.09135008169934641,
            "hi": 0.36666666666666664,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.13636363636363635,
            "p": 0.2546,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "claude-opus-5-5",
          "withProbability": 210,
          "expectedMatches": 8.0,
          "precision": 0.36363636363636365,
          "enrichment": 2.4747474747474745,
          "enrichmentCi": {
            "lo": 1.2478777589134127,
            "hi": 3.9756592292089246,
            "valid": 10000
          },
          "difference": 0.13636363636363635,
          "differenceCi": {
            "lo": -0.07536660241307916,
            "hi": 0.3416776315789472,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.13636363636363635,
            "p": 0.1991,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "claude-haiku-5-5",
          "withProbability": 245,
          "expectedMatches": 6.0,
          "precision": 0.2727272727272727,
          "enrichment": 1.8560606060606057,
          "enrichmentCi": {
            "lo": 0.7,
            "hi": 3.1956521739130435,
            "valid": 10000
          },
          "difference": 0.2272727272727273,
          "differenceCi": {
            "lo": 0.021528537252221355,
            "hi": 0.4523117592789723,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.2272727272727273,
            "p": 0.0433,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.3031,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "claude-opus-4-8",
          "withProbability": 235,
          "expectedMatches": 6.833333333333334,
          "precision": 0.3106060606060606,
          "enrichment": 2.113846801346801,
          "enrichmentCi": {
            "lo": 0.8155263317921733,
            "hi": 3.542900099854757,
            "valid": 10000
          },
          "difference": 0.1893939393939394,
          "differenceCi": {
            "lo": -0.029095581711717,
            "hi": 0.42484370559818124,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.1893939393939394,
            "p": 0.10955,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "gpt-6-astra",
          "withProbability": 240,
          "expectedMatches": 8.5,
          "precision": 0.38636363636363635,
          "enrichment": 2.6294191919191916,
          "enrichmentCi": {
            "lo": 1.011308496070493,
            "hi": 4.276496809309309,
            "valid": 10000
          },
          "difference": 0.11363636363636365,
          "differenceCi": {
            "lo": -0.11111904761904766,
            "hi": 0.36135697700132674,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.11363636363636365,
            "p": 0.345,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "gemini-4-argon-high",
          "withProbability": 245,
          "expectedMatches": 8.0,
          "precision": 0.36363636363636365,
          "enrichment": 2.4747474747474745,
          "enrichmentCi": {
            "lo": 1.1515863689776733,
            "hi": 4.083333333333333,
            "valid": 10000
          },
          "difference": 0.13636363636363635,
          "differenceCi": {
            "lo": -0.025741262414056547,
            "hi": 0.30385869565217377,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.13636363636363635,
            "p": 0.1123046875,
            "draws": 32768,
            "method": "exhaustive paired swap",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        },
        {
          "model": "grok-4.5",
          "withProbability": 220,
          "expectedMatches": 7.0,
          "precision": 0.3181818181818182,
          "enrichment": 2.16540404040404,
          "enrichmentCi": {
            "lo": 0.9072959322959323,
            "hi": 3.8800904977375565,
            "valid": 10000
          },
          "difference": 0.18181818181818182,
          "differenceCi": {
            "lo": -0.040000000000000036,
            "hi": 0.39999999999999997,
            "valid": 10000
          },
          "permutation": {
            "difference": 0.18181818181818182,
            "p": 0.12385,
            "draws": 19999,
            "method": "random paired swaps",
            "pHolm": 0.6573,
            "family": "seven predictors with a numerical probability"
          }
        }
      ]
    },
    "clarityAdjustment": {
      "source": "https://eternalsearch.net/omega/selection-originality-20261030 (manifest, 357 score files) and the two-stage reports",
      "hypotheses": 119,
      "clusters": 58,
      "centre": {
        "clarity": 7.9607843137254894,
        "feasibility": 6.459383753501402
      },
      "outcome": "core article correspondence (majority of three cross-company judges, score 3 or higher)",
      "byRole": {
        "selected": {
          "hypotheses": 60,
          "matches": 37,
          "meanClarity": 8.15,
          "clarityMatched": 8.126126126126128,
          "clarityUnmatched": 8.18840579710145
        },
        "excluded": {
          "hypotheses": 59,
          "matches": 15,
          "meanClarity": 7.768361581920904,
          "clarityMatched": 8.133333333333333,
          "clarityUnmatched": 7.643939393939394
        }
      },
      "models": {
        "selection": {
          "terms": {
            "selected": {
              "coefficient": 1.551563129531126,
              "oddsRatio": 4.718840579710146,
              "robustSe": 0.3528930730084302,
              "waldCi": [
                2.3629071553711203,
                9.423754279173888
              ],
              "p": 1.0991173335744121e-05,
              "bootstrapCi": {
                "lo": 2.5080000000000005,
                "hi": 10.3048128342246,
                "valid": 10000
              },
              "unit": "selected against excluded"
            }
          },
          "bootstrapFailures": 0
        },
        "selection+clarity": {
          "terms": {
            "selected": {
              "coefficient": 1.3483171587156568,
              "oddsRatio": 3.8509395547273693,
              "robustSe": 0.38578120916169384,
              "waldCi": [
                1.807936227158807,
                8.202576634834589
              ],
              "p": 0.0004740072799026076,
              "bootstrapCi": {
                "lo": 1.8965242296898568,
                "hi": 9.03984266899481,
                "valid": 10000
              },
              "unit": "selected against excluded"
            },
            "clarity": {
              "coefficient": 0.7053513136756573,
              "oddsRatio": 2.0245578148477144,
              "robustSe": 0.3408768028847408,
              "waldCi": [
                1.0379346031586039,
                3.949029479494693
              ],
              "p": 0.03852483710911528,
              "bootstrapCi": {
                "lo": 1.0409954237150092,
                "hi": 4.422566533057869,
                "valid": 10000
              },
              "unit": "one point of the mean judge score"
            }
          },
          "bootstrapFailures": 0
        },
        "selection+clarity+feasibility": {
          "terms": {
            "selected": {
              "coefficient": 1.3930100638742824,
              "oddsRatio": 4.026953214605772,
              "robustSe": 0.38708674160425616,
              "waldCi": [
                1.8857394974923456,
                8.599465734364818
              ],
              "p": 0.0003198086553211574,
              "bootstrapCi": {
                "lo": 1.9558654672478462,
                "hi": 9.89959000276041,
                "valid": 10000
              },
              "unit": "selected against excluded"
            },
            "clarity": {
              "coefficient": 0.7602566907770141,
              "oddsRatio": 2.1388251667330325,
              "robustSe": 0.3614439775518637,
              "waldCi": [
                1.053192992173059,
                4.343527850875501
              ],
              "p": 0.03543192317183837,
              "bootstrapCi": {
                "lo": 1.056954590141249,
                "hi": 4.993926248282537,
                "valid": 10000
              },
              "unit": "one point of the mean judge score"
            },
            "feasibility": {
              "coefficient": -0.12785050299784662,
              "oddsRatio": 0.8799849244197276,
              "robustSe": 0.22203570597090316,
              "waldCi": [
                0.5694749479009631,
                1.3598025164412753
              ],
              "p": 0.564743238662226,
              "bootstrapCi": {
                "lo": 0.5567102799385646,
                "hi": 1.4826079105231837,
                "valid": 10000
              },
              "unit": "one point of the mean judge score"
            }
          },
          "bootstrapFailures": 0
        },
        "clarity": {
          "terms": {
            "clarity": {
              "coefficient": 1.013358475557229,
              "oddsRatio": 2.754837550573356,
              "robustSe": 0.29128353258701845,
              "waldCi": [
                1.5565047329923902,
                4.875751270899676
              ],
              "p": 0.0005033977828509027,
              "bootstrapCi": {
                "lo": 1.6219757804292498,
                "hi": 5.414064765078487,
                "valid": 10000
              },
              "unit": "one point of the mean judge score"
            }
          },
          "bootstrapFailures": 0
        }
      },
      "bootstrap": {
        "replicates": 10000,
        "seed": 20261030,
        "unit": "whole task of one generator"
      },
      "pairsInOriginalityReport": 59
    },
    "keyElementsAllSingleQuery": {
      "rule": "an element counts as named when at least two thirds of its content tokens occur in the hypothesis text",
      "hypotheses": 1000,
      "tasks": 50,
      "matches": 156,
      "meanShareMatched": 0.12980769230769232,
      "meanShareUnmatched": 0.04052132701421801,
      "matchedWithNoElementNamed": 84,
      "unmatchedWithAnyElementNamed": 138,
      "auc": 0.652065105116053,
      "aucCi": {
        "lo": 0.6018878312873176,
        "hi": 0.7013368910685932,
        "valid": 10000
      },
      "anyElementNamed": {
        "namedMatched": 72,
        "namedUnmatched": 138,
        "noneMatched": 84,
        "noneUnmatched": 706,
        "oddsRatio": 4.385093167701863,
        "oddsRatioCi": {
          "lo": 2.8737457126904364,
          "hi": 6.814334792786544,
          "valid": 10000
        }
      },
      "byModel": [
        {
          "model": "claude-opus-5-5",
          "hypotheses": 250,
          "matches": 45,
          "auc": 0.6210840108401084
        },
        {
          "model": "gemini-4-argon-high",
          "hypotheses": 250,
          "matches": 33,
          "auc": 0.7616254713028907
        },
        {
          "model": "gpt-6-astra",
          "hypotheses": 250,
          "matches": 36,
          "auc": 0.6292834890965732
        },
        {
          "model": "grok-4.5",
          "hypotheses": 250,
          "matches": 42,
          "auc": 0.6278617216117216
        }
      ],
      "bootstrap": {
        "replicates": 10000,
        "seed": 20261030,
        "unit": "whole task"
      }
    }
  },
  "sourceSha256": {
    "generation-outcomes.json": "b67424c82c8f1a3bc981301d906606c64d271929516b80178a8b9cb9b05d9ddc",
    "https://eternalsearch.net/omega/balanced-opus-20261023/report.json": "3f0914c47ea58754fd7a4c931dbca4d699f5c3619823eb60be4b77c53a71d5d8",
    "https://eternalsearch.net/omega/direct-confirmation-grok-20261022/report.json": "69b2d56db9e11ddab5b1447c9bd9dc24a38d2f229347e910d078265ab7d3e0bc",
    "https://eternalsearch.net/omega/direct-confirmation-grok-20261022/requests.json": "e98331debbc6dec418064fc96e3d4572f8c46c5a7d81d2e4c95b2b6dd7dc55ae",
    "https://eternalsearch.net/omega/experimental-support-20261022/report.json": "ed41799db244cf0bd993f9eb0ec855c43bce9b17c998407a65aef50121ef3dd0",
    "https://eternalsearch.net/omega/gemini-generated-confirmation-20261023/cohort.json": "864112330b7b49bbc4919e1d7a8eee78054581ec16d87a2b8dfcbbdbb2d448f9",
    "https://eternalsearch.net/omega/grok-generated-confirmation-20261023/cohort.json": "dc965b953bb2e17b9da8e357f90d1056da41e94febfb36d8ace049ae1dbf4c3b",
    "https://eternalsearch.net/omega/grok-generated-confirmation-20261023/rediscovery-tasks.json": "068c3136304a39fb31e7a68aa933ce5b4949a46d7f23d5b32fc67c53a688e8f0",
    "https://eternalsearch.net/omega/opus-generated-confirmation-20261023/cohort.json": "2585e4078d9e937f40eb73ff8032f632b2267a65c1b3ba0582a3d2ff6e14e718",
    "https://eternalsearch.net/omega/opus-isolated-confirmation-20261028/cohort.json": "6113d567d36fe001ab364e3538b1e7e7aae62175b7bf9d6933b551ef73d16a6d",
    "https://eternalsearch.net/omega/selection-originality-20261030/manifest.json": "f5f54d5cbb690d4ec97c8b28ead6756ed0150ee1072d4f0bdab1624d843e4e58",
    "https://eternalsearch.net/omega/selection-originality-20261030/report.json": "53266b8fdde4994a6c781a3f5fca33983597d330539cceed54c9d8e411e40869",
    "https://eternalsearch.net/omega/selection-originality-20261030/scores/*.json (name and bytes, sorted)": "0b1451ca86fdf7bf36b30a39c0943d217ccfb699f43b2a317fe5aeee5d74f2c4",
    "https://eternalsearch.net/omega/two-stage-unanimous-20261023/report.json": "e9d7a0e581e184070fa6963458a27253df442d530ecd28e0c5c628c7d0a0f365",
    "https://eternalsearch.net/omega/two-stage-unanimous-gemini-20261023/report.json": "5f0a11da83a66029d3cf0c81f302dddf9ab99da19fe65f2183a66bdf2b253b12",
    "https://eternalsearch.net/omega/two-stage-unanimous-opus-20261023/report.json": "07ed307f6065d2f0b3f3b6eef06416b683a0fb4b4bf2c47c87b819a51bf40cae",
    "https://eternalsearch.net/omega/two-stage-unanimous-opus-isolated-20261028/report.json": "115a9701aa164be1ca5b55176943fa030986e0f08d35dfdcdcc6f4db9419838b"
  }
}
