ResearchDatasetsSWE-SmithFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsSequence-level risk predictionSWE-smith (test)67.6AUC6Watermark DetectionSWE-Smith0.41Delta_q (No Attack)2