NavigationTuringTest(NTT):LearningtoEvaluateHuman-LikeNavigationSamDevlin1RalucaGeorgescu1IdaMomennejad2JaroslawRzepecki1EvelynZuniga1GavinCostello3GuyLeroy1AliShaw3KatjaHofmann1AbstractmatedNTTorA...
LEEP:ANewMeasuretoEvaluateTransferabilityofLearnedRepresentationsCuongV.Nguyen1TalHassner2MatthiasSeeger1CedricArchambeau1Abstractchoosegoodsourcemodelsforagiventargettask(Achilleetal.,2019;Baoetal...
DeepValueNetworksLearntoEvaluateandIterativelyRefineStructuredOutputsMichaelGygli1MohammadNorouzi2AneliaAngelova2Abstractcomplicatedhighlevelreasoningtoresolveambiguity.Weapproachstructuredoutputpr...