ResearchBenchmarksResponse Helpfulness on SOTOPIA AURATownFollow3.83Average Human RatingAURA2.18682.61343.043.4666Jun 4, 2026Evaluation ResultsMethodMethodLinksAverage Human RatingDelta Mean Rater MeansWilcoxon p-value (n=8)95% Cluster CI (Lower Bound)AURARating Scale=5-point L...Rating Scale=5-point Likert, Raters=8, Scenarios=50, Warmup=10 steps, Seed=422026.063.831.580.0170.94VanillaRating Scale=5-point L...Rating Scale=5-point Likert, Raters=8, Scenarios=50, Warmup=10 steps, Seed=422026.062.25———