Multi-domain Task Incremental Learning on MTIL Last 1.0 (test)
62Accuracy (Aircraft)Individual FT
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Individual FTFine-tuning Strategy=FT2025.03 | 62 | 95.1 | 89.6 | 79.5 | 98.9 | 97.5 | 92.7 | 99.6 | 94.7 | 89.6 | 81.8 | 89.2 | |
| ConDUFine-tuning Strategy=FT, Training Paradigm=Last2025.03 | 58.6 | 93.7 | 86.6 | 76.1 | 98.2 | 93.4 | 91.9 | 99.6 | 94.8 | 84.9 | 80.5 | 87.1 | |
| Dual-RAILTraining Paradigm=Last2025.03 | 52.5 | 96.8 | 83.3 | 80.1 | 96.4 | 99 | 89.9 | 98.8 | 93.5 | 85.5 | 79.2 | 86.8 | |
| DPeCLIPTraining Paradigm=Last2025.03 | 49.9 | 95.6 | 85.8 | 78.6 | 98.4 | 95.8 | 92.1 | 99.4 | 94 | 84.5 | 81.7 | 86.9 | |
| MulKITraining Paradigm=Last2025.03 | 49.7 | 93 | 82.8 | 73.7 | 96.2 | 92.3 | 90.4 | 99 | 94.8 | 85.2 | 78.9 | 85.1 | |
| ConDUFine-tuning Strategy=LoRA, Training Paradigm=Last2025.03 | 48.9 | 95.2 | 87.8 | 78.5 | 96.3 | 95.2 | 91.7 | 97.6 | 93 | 85.3 | 78.8 | 86.2 | |
| ZSCLTraining Paradigm=Last2025.03 | 40.6 | 92.2 | 81.3 | 70.5 | 94.8 | 90.5 | 91.9 | 98.7 | 93.9 | 85.3 | 80.2 | 83.6 | |
| Zero-shotFine-tuning Strategy=None2025.03 | 24.3 | 88.4 | 68.2 | 44.6 | 54.9 | 71 | 88.5 | 59.4 | 89 | 64.7 | 65.2 | 65.3 |