ResearchDatasetsRadNLIRadNLIFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSummarizationRadNLI39.6ROUGE-L12Natural Language InferenceRadNLI83.3F1 Score12Sentence SimilarityRadNLI (145 pairs)89.66Accuracy3