{ "configurations": {}, "experiment": "llm-authored-rare-intent-supplement-v1", "results": [ { "authoredDeployedThresholds": { "macroF1": 0.3415, "macroPrecision": 0.3925, "macroRecall": 0.4449, "records": 144 }, "authoredRecalibratedThresholds": { "macroF1": 0.3864, "macroPrecision": 0.4341, "macroRecall": 0.448, "records": 144 }, "baseline": { "macroF1": 0.2651, "macroPrecision": 0.418, "macroRecall": 0.2798, "records": 144 }, "corpus": "authoredEvaluation" }, { "authoredDeployedThresholds": { "macroF1": 0.7138, "macroPrecision": 0.8635, "macroRecall": 0.6273, "records": 680 }, "authoredRecalibratedThresholds": { "macroF1": 0.6978, "macroPrecision": 0.8722, "macroRecall": 0.6023, "records": 680 }, "baseline": { "macroF1": 0.6817, "macroPrecision": 0.874, "macroRecall": 0.5829, "records": 680 }, "corpus": "randomHoldout" }, { "authoredDeployedThresholds": { "macroF1": 0.4166, "macroPrecision": 0.525, "macroRecall": 0.3558, "records": 640 }, "authoredRecalibratedThresholds": { "macroF1": 0.4166, "macroPrecision": 0.525, "macroRecall": 0.3558, "records": 640 }, "baseline": { "macroF1": 0.4166, "macroPrecision": 0.525, "macroRecall": 0.3558, "records": 640 }, "corpus": "targetedReleaseHoldout" }, { "authoredDeployedThresholds": { "macroF1": 0.1486, "macroPrecision": 0.3363, "macroRecall": 0.1116, "records": 200 }, "authoredRecalibratedThresholds": { "macroF1": 0.1486, "macroPrecision": 0.3363, "macroRecall": 0.1116, "records": 200 }, "baseline": { "macroF1": 0.1486, "macroPrecision": 0.3363, "macroRecall": 0.1116, "records": 200 }, "corpus": "onlineRealHoldout" }, { "authoredDeployedThresholds": { "macroF1": 0.2726, "macroPrecision": 0.291, "macroRecall": 0.3919, "records": 41195 }, "authoredRecalibratedThresholds": { "macroF1": 0.2744, "macroPrecision": 0.2919, "macroRecall": 0.3834, "records": 41195 }, "baseline": { "macroF1": 0.2783, "macroPrecision": 0.2948, "macroRecall": 0.332, "records": 41195 }, "corpus": "comprehensiveOnlineHoldout" } ], "schemaVersion": 1 }