ResearchDatasetsWerewolfFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsSocial Deduction Game PlayWerewolf against human players (test)3.21Average Votes7Multi-agent interaction and social reasoningWerewolf MultiAgentBench55.75Task Performance6Social Deduction Game GameplayWerewolf29.1Village Team Win Rate6Dimensional Emotion RecognitionWerewolf-XL39.8Arousal5