List your company
ResearchFormerly WizWand

Machine Comprehension on CBT NE (test)

81.6AccuracyHumans
40.20850.95461.772.446Mar 4, 2016Apr 14, 2016May 25, 2016Jul 5, 2016Aug 15, 2016Sep 25, 2016Nov 6, 2016
Updated 8mo ago

Evaluation results

MethodLinks
Humanscontext + query=true2016.0681.6
HumansInput configuration=context+query2016.0381.6
Humansinput=context+query2016.0681.6
Humans (context + query)2016.0681.6
AS Reader (+BookTest, ensemble)trained_on_large_set=true, ensemble=true2016.0678.4
AS Reader (+BookTest)trained_on_large_set=true2016.0676.2
FG fine-grained gateArchitecture=FG2016.1174.96
GA Readervariant=GA (+feature, fix L(w))2016.0674.9
GA fine-grained gateArchitecture=GA2016.1174.64
GA Readervariant=GA (+feature, update L(w))2016.0673.3
NSE2016.0673.2
Munkhdalai & Yu (2016)2016.1173.2
GA scalar gateArchitecture=GA2016.1172.6
GA word char feat concatArchitecture=GA2016.1172.56
GA word char concatArchitecture=GA2016.1172.5
Iterative Alternating Neural Attentionensemble=true2016.0672
AoA Reader2016.0672
Iterative Attentive Readerensemble=true2016.0672
Cui et al. (2016)2016.1172
Sordoni et al. (2016) ensembleEnsemble=true2016.1172
EpiReaderensemble=true2016.0671.8
EpiReaderensemble=true2016.0671.8
Trischler et al. (2016) ensembleEnsemble=true2016.1171.8
GA Readervariant=GA (fix L(w))2016.0671.4
AS ReaderModel variant=greedy ensemble2016.0371
AS Readerensemble=true2016.0671
Kadlec et al. (2016) ensembleEnsemble=true2016.1171
AS Readerensemble=true2016.0670.6
AS ReaderModel variant=avg ensemble2016.0370.6
AS Readerensemble=true2016.0670.6
GA Readervariant=GA (update L(w))2016.0670.1
EpiReader2016.0669.7
EpiReader2016.0669.7
Trischler et al. (2016)2016.1169.7
EpiReader Extractor2016.0669.4
GA Readervariant=GA--2016.0669
AS Reader2016.0668.6
AS ReaderModel variant=single model2016.0368.6
AS Reader2016.0668.6
Iterative Alternating Neural Attention2016.0668.6
AS Reader2016.0668.6
Iterative Attentive Reader2016.0668.6
Sordoni et al. (2016)2016.1168.6
AS ReaderModel variant=avg for top 20%2016.0368.4
MemNNs2016.0666.6
MemNNsModel variant=window memory + self-sup.2016.0366.6
MemNNsconfig=window memory + self-sup.2016.0666.6
MemNets2016.0666.6
Iterative Alternating Neural Attentionattention=fixed query attention2016.0666
HumansInput configuration=query2016.0352
Humansinput=query2016.0652
Humans (query)2016.0652
LSTMscontext + query=true2016.0641.8
LSTMsInput configuration=context+query2016.0341.8
LSTMsinput=context+query2016.0641.8
LSTMS (context + query)2016.0641.8