Continual Learning on FOMC
59Average AccuracySETA
Evaluation results
| Method | Links | |
|---|---|---|
| SETABackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 59 | |
| I-LoRABackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 53.9 | |
| EWCBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 53 | |
| GEMBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 46.4 | |
| ERBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 45.6 | |
| A-GEMBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 43.9 | |
| PPBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 42.1 | |
| Seq-TrainBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 41.1 | |
| L2PBackbone=LLaMA-2 7B, Batch size=16, Learning rate=1e-42026.01 | 39.5 |