Text Classification On Yelp 5
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||
|---|---|---|---|
| HAHNN (CNN) | 73.28% | Hierarchical Attentional Hybrid Neural Networks for Document Classification | |
| XLNet | 72.95% | XLNet: Generalized Autoregressive Pretraining for Language Understanding | |
| BigBird | 72.16% | Big Bird: Transformers for Longer Sequences | |
| BERT-ITPT-FiT | 70.58% | How to Fine-Tune BERT for Text Classification? | |
| LSTM-reg (single moedl) | 68.7% | Rethinking Complex Neural Network Architectures for Document Classification | - |
| BERT Finetune + UDA | 67.92% | Unsupervised Data Augmentation for Consistency Training | |
| ULMFiT (Small data) | 67.6% | Sampling Bias in Deep Active Classification: An Empirical Study |
0 of 7 row(s) selected.