ResearchTasksProbability ForecastingProbability ForecastingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedPolymarket 30% held-out evaluation cliqueshierarchical JCD0.141Brier Score20May 29, 2026Paleka + Polymarket partition slice N=2412 (combined)meta_llama-3.3-70b-versatile0.249B2 Score4May 29, 2026
Paleka + Polymarket partition slice N=2412 (combined)meta_llama-3.3-70b-versatile0.249B2 Score4May 29, 2026