ResearchDatasetsBT-guidedBT-guidedFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsExpected Calibration ErrorBT-guided2ECE (15-bin)9Preference ModelingBT-guided1.09NLL8Multiple choice question answeringBT-guided (test)38.4Accuracy7