{
  "schema_version": 1,
  "columns": [
    "configuration",
    "bin",
    "score_min_observed",
    "score_max_observed",
    "species_attempts",
    "species_correct",
    "species_wrong",
    "binning",
    "estimate",
    "ci_low",
    "ci_high",
    "valid_resamples",
    "interval_status"
  ],
  "rows": [
    {
      "configuration": "BioCLIP 2.5 / primary_blind",
      "bin": "0",
      "score_min_observed": "0.051088",
      "score_max_observed": "0.4254",
      "species_attempts": "158",
      "species_correct": "11",
      "species_wrong": "147",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.06962025316455696",
      "ci_low": "0.03225806451612903",
      "ci_high": "0.11180124223602485",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / primary_blind",
      "bin": "1",
      "score_min_observed": "0.426345",
      "score_max_observed": "0.659248",
      "species_attempts": "158",
      "species_correct": "28",
      "species_wrong": "130",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.17721518987341772",
      "ci_low": "0.11842105263157894",
      "ci_high": "0.23841059602649006",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / primary_blind",
      "bin": "2",
      "score_min_observed": "0.662144",
      "score_max_observed": "0.910807",
      "species_attempts": "158",
      "species_correct": "40",
      "species_wrong": "118",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.25316455696202533",
      "ci_low": "0.1858974358974359",
      "ci_high": "0.3248407643312102",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / primary_blind",
      "bin": "3",
      "score_min_observed": "0.914371",
      "score_max_observed": "0.997865",
      "species_attempts": "158",
      "species_correct": "89",
      "species_wrong": "69",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.5632911392405063",
      "ci_low": "0.4863013698630137",
      "ci_high": "0.6413825634352635",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / primary_blind",
      "bin": "4",
      "score_min_observed": "0.997905",
      "score_max_observed": "1.0",
      "species_attempts": "158",
      "species_correct": "154",
      "species_wrong": "4",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.9746835443037974",
      "ci_low": "0.9473684210526315",
      "ci_high": "0.9941520467836257",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / primary_blind",
      "bin": "0",
      "score_min_observed": "0.6501",
      "score_max_observed": "0.794",
      "species_attempts": "77",
      "species_correct": "27",
      "species_wrong": "50",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.35064935064935066",
      "ci_low": "0.24324324324324326",
      "ci_high": "0.4657534246575342",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / primary_blind",
      "bin": "1",
      "score_min_observed": "0.7965",
      "score_max_observed": "0.9131",
      "species_attempts": "76",
      "species_correct": "31",
      "species_wrong": "45",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.40789473684210525",
      "ci_low": "0.29850746268656714",
      "ci_high": "0.5194805194805194",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / primary_blind",
      "bin": "2",
      "score_min_observed": "0.9151",
      "score_max_observed": "0.9724",
      "species_attempts": "77",
      "species_correct": "41",
      "species_wrong": "36",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.5324675324675324",
      "ci_low": "0.4133174603174603",
      "ci_high": "0.6511627906976745",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / primary_blind",
      "bin": "3",
      "score_min_observed": "0.9725",
      "score_max_observed": "0.9947",
      "species_attempts": "76",
      "species_correct": "48",
      "species_wrong": "28",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.631578947368421",
      "ci_low": "0.5066666666666667",
      "ci_high": "0.7528089887640449",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / primary_blind",
      "bin": "4",
      "score_min_observed": "0.9948",
      "score_max_observed": "1.0",
      "species_attempts": "77",
      "species_correct": "71",
      "species_wrong": "6",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.922077922077922",
      "ci_low": "0.855072463768116",
      "ci_high": "0.9759036144578314",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / operational_country_ablation",
      "bin": "0",
      "score_min_observed": "0.140242",
      "score_max_observed": "0.521949",
      "species_attempts": "158",
      "species_correct": "14",
      "species_wrong": "144",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.08860759493670886",
      "ci_low": "0.04666666666666667",
      "ci_high": "0.13548387096774195",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / operational_country_ablation",
      "bin": "1",
      "score_min_observed": "0.522083",
      "score_max_observed": "0.768456",
      "species_attempts": "158",
      "species_correct": "32",
      "species_wrong": "126",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.20253164556962025",
      "ci_low": "0.14093959731543623",
      "ci_high": "0.267515923566879",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / operational_country_ablation",
      "bin": "2",
      "score_min_observed": "0.769699",
      "score_max_observed": "0.961624",
      "species_attempts": "158",
      "species_correct": "42",
      "species_wrong": "116",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.26582278481012656",
      "ci_low": "0.19718052842603062",
      "ci_high": "0.33544303797468356",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / operational_country_ablation",
      "bin": "3",
      "score_min_observed": "0.961788",
      "score_max_observed": "0.999302",
      "species_attempts": "158",
      "species_correct": "105",
      "species_wrong": "53",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.6645569620253164",
      "ci_low": "0.5862014056665934",
      "ci_high": "0.7409649388922597",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "BioCLIP 2.5 / operational_country_ablation",
      "bin": "4",
      "score_min_observed": "0.999327",
      "score_max_observed": "1.0",
      "species_attempts": "158",
      "species_correct": "154",
      "species_wrong": "4",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.9746835443037974",
      "ci_low": "0.9464255752636624",
      "ci_high": "0.9941860465116279",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / operational_country_ablation",
      "bin": "0",
      "score_min_observed": "0.6559",
      "score_max_observed": "0.8492",
      "species_attempts": "55",
      "species_correct": "39",
      "species_wrong": "16",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.7090909090909091",
      "ci_low": "0.5869377811094454",
      "ci_high": "0.8253968253968254",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / operational_country_ablation",
      "bin": "1",
      "score_min_observed": "0.8497",
      "score_max_observed": "0.9389",
      "species_attempts": "55",
      "species_correct": "38",
      "species_wrong": "17",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.6909090909090909",
      "ci_low": "0.5609663710273467",
      "ci_high": "0.8163265306122449",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / operational_country_ablation",
      "bin": "2",
      "score_min_observed": "0.9391",
      "score_max_observed": "0.985",
      "species_attempts": "55",
      "species_correct": "45",
      "species_wrong": "10",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.8181818181818182",
      "ci_low": "0.7068965517241379",
      "ci_high": "0.9166666666666666",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / operational_country_ablation",
      "bin": "3",
      "score_min_observed": "0.9854",
      "score_max_observed": "0.997",
      "species_attempts": "52",
      "species_correct": "45",
      "species_wrong": "7",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.8653846153846154",
      "ci_low": "0.7413633461047254",
      "ci_high": "0.9736842105263158",
      "valid_resamples": "10000",
      "interval_status": "ok"
    },
    {
      "configuration": "SpeciesNet v4.0.3b / operational_country_ablation",
      "bin": "4",
      "score_min_observed": "0.9974",
      "score_max_observed": "1.0",
      "species_attempts": "58",
      "species_correct": "57",
      "species_wrong": "1",
      "binning": "within-route empirical quintiles; ties kept together; exploratory",
      "estimate": "0.9827586206896551",
      "ci_low": "0.9411764705882353",
      "ci_high": "1.0",
      "valid_resamples": "10000",
      "interval_status": "ok"
    }
  ],
  "unit_policy": {
    "table": "specialist_native_score_reliability.csv",
    "metric_units": {},
    "counts": {
      "species_attempts": "answers",
      "species_correct": "answers",
      "species_wrong": "answers",
      "valid_resamples": "bootstrap_resamples"
    },
    "interval": "ci_low and ci_high use the same units as estimate; pointwise 95% when interval_status is ok",
    "estimate": "proportion_0_to_1 conditional species correctness within a native-score bin",
    "score_min_observed": "route_native_score_not_calibrated_probability",
    "score_max_observed": "route_native_score_not_calibrated_probability",
    "scope": "within-route empirical bins only; do not compare native scores across models"
  },
  "note": "Saved aggregate candidate analysis; intervals are pointwise and exploratory tests have separate Holm-adjusted p-values."
}
