ResearchDatasetsHeld-out observerHeld-out observerFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA 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