Natural Language Inference On Farstail
Metrics
% Test Accuracy
Results
Performance results of various models on this benchmark
Model Name | % Test Accuracy | Paper Title | Repository |
|---|---|---|---|
| HBMP + word2vec | 66.04 | FarsTail: A Persian Natural Language Inference Dataset | |
| Translate-Source + fastText | 78.13 | FarsTail: A Persian Natural Language Inference Dataset | |
| ESIM + fastText | 71.16 | FarsTail: A Persian Natural Language Inference Dataset | |
| LSTM + BERT (concat) | 75.83 | FarsTail: A Persian Natural Language Inference Dataset | |
| Decomposable Attention Model + word2vec | 66.62 | FarsTail: A Persian Natural Language Inference Dataset | |
| Translate-Target + fastText | 70.46 | FarsTail: A Persian Natural Language Inference Dataset | |
| mBERT | 83.38 | FarsTail: A Persian Natural Language Inference Dataset | |
| ULMFiT | 72.44 | FarsTail: A Persian Natural Language Inference Dataset | |
| ParsBERT | 82.99 | FarsTail: A Persian Natural Language Inference Dataset | |
| ESIM + BERT (FarsTail, MultiNLI) | 74.62 | FarsTail: A Persian Natural Language Inference Dataset |
0 of 10 row(s) selected.