ResearchDatasetsHeld-out observerFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsSystem-effect predictionHeld-out observer system-effect Non-endpoint features (source rows evals)93.57Accuracy1
System-effect predictionHeld-out observer system-effect Non-endpoint features (source rows evals)93.57Accuracy1