Aspect Extraction and Sentiment Classification on 14res (test)
82.6F1 ScoreAESC-Contra
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AESC-ContraBackbone=T5, Contrastive Pre-training=true2023.10 | 82.6 | — | — | |
| AESC-BaseBackbone=T5, Contrastive Pre-training=false2023.10 | 81.8 | — | — | |
| SentiPrompt2023.10 | 81.1 | — | — | |
| EHG2023.10 | 79.3 | — | — | |
| GAS-R2023.10 | 79 | — | — | |
| Unified-BART-ABSA2023.10 | 78.5 | — | — | |
| Dual-MRCBackbone=BERT-Base-Uncased2021.01 | 76.57 | 76.84 | 76.31 | |
| Peng-two-stageTagging Schema=unified tagging2021.01 | 74.19 | 74.41 | 73.97 | |
| Li-unified-RTagging Schema=unified tagging2021.01 | 73.79 | 73.15 | 74.44 | |
| CMLA2021.01 | 70.62 | 67.8 | 73.69 | |
| ChatGPT2023.10 | 63.4 | — | — | |
| RINANTE2021.01 | 48.15 | 48.97 | 47.36 |