ResearchBenchmarksText Classification on MultiNLIFollow81.1Average AccuracyGroupDRO68.51671.78375.0578.317Nov 6, 2025Evaluation ResultsMethodMethodLinksAverage AccuracyWorst AccuracyGroupDRO2025.1181.176ERM2025.1180.966.8SCER2025.1180.476.8LISA2025.1180.373.3ReSample2025.1177.272.3PDE2025.1169.165.8ElRep2025.116966.8