List your company
ResearchFormerly WizWand

Sentence Classification on Stanford Sentiment Treebank (SST) fine-grained (test)

53.1AccuracyNTI-SLSTM-LSTM
43.386445.908248.4350.9518Jul 15, 2016Jan 6, 2017Jun 30, 2017Dec 22, 2017Jun 15, 2018Dec 7, 2018May 31, 2019
Updated 8mo ago

Evaluation results

MethodLinks
NTI-SLSTM-LSTMleaf_function=SLSTM, composition_function=LSTM2016.0753.1
DMN2016.0752.1
DiSANDirections=true2017.0951.72
CNN-Tensor2017.0951.2
NCSL2017.0951.1
CT-LSTMtype=Constituency Tree2016.0751
Tree-LSTM2017.0951
LR-Bi-LSTM2017.0950.6
NTI-SLSTMleaf_function=SLSTM2016.0750.5
Bi-LSTM with s2t self-attentionBackbone=Bi-LSTM, Attention=s2t self-attention2017.0949.95
NLM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=CNN2019.0549.9
CNN-400Emb.Size=4002019.0549.86
LM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=CNN2019.0549.82
DRNN2016.0749.8
Bi-LSTM2017.0949.8
DiSAN without directionsDirections=false2017.0949.41
LM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=LSTM2019.0549.37
LSTM-400Emb.Size=4002019.0549.23
STM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=CNN2019.0549.19
Multi-head with s2t self-attentionAttention=multi-head s2t self-attention2017.0949.14
STM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=LSTM2019.0549.14
Bi-LSTMbidirectional=true2016.0749.1
STM+TSEDEmb.Size=50, Teacher Model Architecture=CNN2019.0549.1
Word Embedding with s2t self-attentionAttention=s2t self-attention2017.0948.87
NLM+TSED+PT+2LEmb.Size=50, Teacher Model Architecture=LSTM2019.0548.87
NLM+TSEDEmb.Size=50, Teacher Model Architecture=LSTM2019.0548.81
LM+TSEDEmb.Size=50, Teacher Model Architecture=LSTM2019.0548.79
STM+TSEDEmb.Size=50, Teacher Model Architecture=LSTM2019.0548.77
LM+TSEDEmb.Size=50, Teacher Model Architecture=CNN2019.0548.71
NLM+TSEDEmb.Size=50, Teacher Model Architecture=CNN2019.0548.66
CNN-non-static2017.0948
Word Embedding with additive attentionAttention=additive2017.0947.47
CNN-MC2016.0747.4
2-layer LSTMlayers=22016.0746
RNTN2016.0745.7
RNTN2017.0945.7
CNN-50Emb.Size=502019.0545.07
ENCEmb.Size=502019.0544.94
MV-RNN2017.0944.4
LSTM-50Emb.Size=502019.0543.76