Aspect Sentiment Classification on ASC
91.91AccuracyMTL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MTLScenario=Non-continual Learning (SDL), Backbone=BERT2021.12 | 91.91 | 88.11 | |
| CTRScenario=Continual Learning (CL), Backbone=BERT2021.12 | 89.47 | 83.62 | |
| LAMOLScenario=Continual Learning (CL), Backbone=GPT-22021.12 | 88.91 | 80.59 | |
| B-CLScenario=Continual Learning (CL), Backbone=BERT2021.12 | 88.29 | 81.4 | |
| CTR (forward)Scenario=Continual Learning (CL), Backbone=BERT2021.12 | 87.89 | 80.25 | |
| OWMScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 87.02 | 79.31 | |
| HATScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 86.74 | 78.16 | |
| EWCScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 86.37 | 74.52 | |
| HATScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 86.14 | 78.52 | |
| A-GEMScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 86.06 | 78.44 | |
| SDLScenario=Non-continual Learning (SDL), Backbone=Adapter-BERT2021.12 | 85.96 | 78.07 | |
| SDLScenario=Non-continual Learning (SDL), Backbone=BERT2021.12 | 85.84 | 76.35 | |
| NFHScenario=Continual Learning (CL), Backbone=BERT (Frozen)2021.12 | 85.51 | 76.64 | |
| KANScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 85.49 | 77.38 | |
| SRKScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 84.76 | 78.52 | |
| UCLScenario=Continual Learning (CL), Backbone=W2V2021.12 | 84.41 | 75.99 | |
| DER++Scenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 84.27 | 75.08 | |
| EWCScenario=Continual Learning (CL), Backbone=W2V2021.12 | 84.16 | 72.29 | |
| UCLScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 83.89 | 74.82 | |
| CATScenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 83.68 | 68.64 | |
| DER++Scenario=Continual Learning (CL), Backbone=W2V2021.12 | 83.27 | 69.93 | |
| OWMScenario=Continual Learning (CL), Backbone=W2V2021.12 | 82.7 | 71.18 | |
| NFHScenario=Continual Learning (CL), Backbone=W2V2021.12 | 82.69 | 73.56 | |
| A-GEMScenario=Continual Learning (CL), Backbone=W2V2021.12 | 81.33 | 63.35 | |
| HATScenario=Continual Learning (CL), Backbone=W2V2021.12 | 80.83 | 63.63 | |
| SDLScenario=Non-continual Learning (SDL), Backbone=BERT (Frozen)2021.12 | 78.14 | 58.13 | |
| SDLScenario=Non-continual Learning (SDL), Backbone=W2V2021.12 | 77.01 | 51.89 | |
| CATScenario=Continual Learning (CL), Backbone=W2V2021.12 | 76.28 | 54.65 | |
| OWMScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 72.99 | 66.51 | |
| KANScenario=Continual Learning (CL), Backbone=W2V2021.12 | 72.06 | 40.01 | |
| SRKScenario=Continual Learning (CL), Backbone=W2V2021.12 | 71.01 | 39.63 | |
| UCLScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 64.46 | 36.64 | |
| L2Scenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 63.97 | 52.43 | |
| L2Scenario=Continual Learning (CL), Backbone=W2V2021.12 | 60.36 | 39.13 | |
| EWCScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 56.3 | 49.58 | |
| L2Scenario=Continual Learning (CL), Backbone=BERT (frozen)2021.12 | 56.04 | 38.4 | |
| NFHScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 54.03 | 44.81 | |
| NFHScenario=Continual Learning (CL), Backbone=BERT2021.12 | 49.6 | 43.08 | |
| DER++Scenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 47.63 | 35.54 | |
| A-GEMScenario=Continual Learning (CL), Backbone=Adapter-BERT2021.12 | 45.88 | 28.21 |