ResearchBenchmarksScientific Data Analysis on MLE-bench AI4Science mix of seen and unseenScientific Data Analysis on MLE-bench AI4Science mix of seen and unseenFollow100Stanford COVID Vaccine ScoreAIDE9597.5100102.5Jan 9, 2026Evaluation resultsMethodMethodLinksStanford COVID Vaccine ScoreStatoil Iceberg Classifier AccuracyVentilator Pressure Prediction ScoreAerial Cactus Identification AccuracyHistopathologic Cancer Detection AccuracyAverage Beat RatioAIDEstatus=Baseline, indep...status=Baseline, independent runs=32026.0110047.530.869.899.269.5ForeAgentstatus=Ours, independe...status=Ours, independent runs=32026.0110053.129.587.799.273.9