Domain-Incremental Learning on Office-Home
88.25Last AccuracyE2-LoRA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| E2-LoRABackbone=ViT-B/16-IN1K2026.05 | 88.25 | — | 84.94 | |
| DUCTBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 86.91 | 86.27 | — | |
| DUCTBackbone=ViT-B/16-IN1K2026.05 | 85.42 | — | 81.28 | |
| CODA-PromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 85.07 | 84.7 | — | |
| CODA-PromptBackbone=ViT-B/16-IN1K2026.05 | 85.07 | — | 84.7 | |
| DCEBackbone=ViT-B/16-IN1K2026.05 | 84.4 | — | 84.6 | |
| ReplayBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10 | 83.75 | 84.23 | — | |
| RanPACBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 82.28 | 82.3 | — | |
| RanPACBackbone=ViT-B/16-IN1K2026.05 | 82.28 | — | 82.3 | |
| iCaRLBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10 | 81.11 | 81.66 | — | |
| DualPromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 80.85 | 80.2 | — | |
| DualPromptBackbone=ViT-B/16-IN1K2026.05 | 80.85 | — | 80.2 | |
| S-iPromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 80.51 | 81.5 | — | |
| L2PBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 80.03 | 79.72 | — | |
| L2PBackbone=ViT-B/16-IN1K2026.05 | 80.03 | — | 79.72 | |
| EASEBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 76.33 | 81.16 | — | |
| EASEBackbone=ViT-B/16-IN1K2026.05 | 76.33 | — | 81.16 | |
| FinetuneBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 76.16 | 78.32 | — | |
| SimpleCILBackbone=ViT-B/16-IN1K, Exemplars=02024.10 | 75.72 | 75.69 | — | |
| SimpleCILBackbone=ViT-B/16-IN1K2026.05 | 75.72 | — | 75.69 | |
| MEMOBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10 | 63.09 | 71.18 | — | |
| MEMOBackbone=ViT-B/16-IN1K2026.05 | 63.09 | — | 71.18 | |
| AvgMemory=Θ(N)2025.09 | — | 85.21 | — | |
| CMMMemory=Θ(1)2025.09 | — | 89.29 | — | |
| EWCMemory=Θ(1)2025.09 | — | 88.06 | — | |
| LwFMemory=Θ(1)2025.09 | — | 88.56 | — | |
| MaxAbsMemory=Θ(N)2025.09 | — | 89.09 | — | |
| RandMixMemory=Θ(N)2025.09 | — | 89.45 | — | |
| TIESMemory=Θ(N)2025.09 | — | 88.59 | — |