Multiclass text classification on Multilingual Amazon Reviews Corpus (test)
93.5Accuracy (Avg)RETVec
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| RETVecModel=RNN2023.02 | 93.5 | — | — | — | — | |
| WhitespaceModel=RNN2023.02 | 93.2 | — | — | — | — | |
| WhitespaceModel=CNN2023.02 | 93.2 | — | — | — | — | |
| RETVec-rawModel=RNN2023.02 | 92.9 | — | — | — | — | |
| RETVec-rawModel=BERT2023.02 | 92.7 | — | — | — | — | |
| RETVecModel=BERT2023.02 | 92.6 | — | — | — | — | |
| RETVecModel=CNN2023.02 | 92.2 | — | — | — | — | |
| WhitespaceModel=BERT2023.02 | 91.8 | — | — | — | — | |
| RETVec-rawModel=CNN2023.02 | 91.1 | — | — | — | — | |
| SentencePieceModel=CNN2023.02 | 90.8 | — | — | — | — | |
| SentencePieceModel=RNN2023.02 | 90.7 | — | — | — | — | |
| BPEModel=RNN2023.02 | 90.3 | — | — | — | — | |
| BPEModel=CNN2023.02 | 89.2 | — | — | — | — | |
| SentencePieceModel=BERT2023.02 | 87.3 | — | — | — | — | |
| BPEModel=BERT2023.02 | 87.2 | — | — | — | — | |
| fastTextModel=RNN2023.02 | 87 | — | — | — | — | |
| fastTextModel=BERT2023.02 | 86.6 | — | — | — | — | |
| fastTextModel=CNN2023.02 | 84.4 | — | — | — | — | |
| LS-unLLaMAEvaluation Setting=label-supervised adaptation2023.10 | 59.21 | 56.9 | 60.2 | 49.68 | 69.7 | |
| LS-LLaMAEvaluation Setting=label-supervised adaptation2023.10 | 58.41 | 56.8 | 58.82 | 49.28 | 68.72 | |
| BERT-Base-MultilingualEvaluation Setting=Discriminant baselines2023.10 | 55.13 | 53.54 | 53.42 | 46.22 | 67.35 | |
| RoBERTa-Large-MultilingualEvaluation Setting=Discriminant baselines2023.10 | 54.52 | 52.76 | 53.64 | 44.76 | 66.9 | |
| RoBERTa-Base-MultilingualEvaluation Setting=Discriminant baselines2023.10 | 54.12 | 52.61 | 52.56 | 44.32 | 66.98 | |
| LLaMA-2-7BEvaluation Setting=Zero-shot2023.10 | 12.11 | 9.23 | 12.13 | 7.43 | 19.65 |