Natural Language Inference on MultiNLI mismatched (dev)
88.4AccuracyaMLP_large
Evaluation results
| Method | Links | |
|---|---|---|
| aMLP_largeAttn Size=128, Params (M)=3162021.05 | 88.4 | |
| gMLP_xlargeParams (M)=9412021.05 | 87.7 | |
| BERT_large (ours)Attn Size=1024 (64 x 16), Params (M)=3362021.05 | 87.4 | |
| gMLP_largeParams (M)=3652021.05 | 86.5 | |
| aMLP_baseAttn Size=64, Params (M)=1092021.05 | 85.8 | |
| BERT_base (ours)Attn Size=768 (64 x 12), Params (M)=1102021.05 | 85.7 | |
| gMLP_baseParams (M)=1302021.05 | 84.1 | |
| ESIM GloVeAuxiliary Information=GloVe vectors2017.06 | 73.32 | |
| ESIM dictionaryAuxiliary Information=dictionary definitions2017.06 | 71.65 | |
| ESIM spellingAuxiliary Information=spelling2017.06 | 70.48 | |
| ESIM baselineAuxiliary Information=none2017.06 | 67.22 |