ResearchDatasetsBlackjackFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsBlackjackBlackjack 900 games4,090Win Rate24VLM ReasoningBlackJack (In-Distribution)43.8Success Rate15Blackjack Decision Policy EvaluationBlackjack (Out-of-Sample)46.3Hit-Rate5Blackjack Decision Policy EvaluationBlackjack (In-Sample)44.3Hit-Rate5Card GamesBlackJack42.6SR3