List your company
ResearchFormerly WizWand

Domain-Incremental Learning on CDDB Hard

89.1Accuracy (AT)CoVON
49.049659.447369.84580.2427Jun 22, 2026
Updated 4mo ago

Evaluation results

MethodLinks
CoVONModel=CLIP with pretrained ViT-B/16 as vision encoder2026.0689.1——0.56
MoP-CLIPModel=CLIP with pretrained ViT-B/16 as vision encoder2026.0688.65——0.69
DyToxModel=Vision Transformer (ViT-B/16)2026.0686.21——1.55
S-liPromptsModel=CLIP with pretrained ViT-B/16 as vision encoder2026.0686.08——1.12
PINAModel=CLIP with pretrained ViT-B/16 as vision encoder2026.0685.71——1.21
CoVONModel=Vision Transformer (ViT-B/16)2026.0682.24——0.47
CoVON-NoMModel=Vision Transformer (ViT-B/16)2026.0680.11——0.67
PINAModel=Vision Transformer (ViT-B/16)2026.0677.35——0.98
S-iPromptsModel=Vision Transformer (ViT-B/16)2026.0674.51——1.3
CODA-PModel=Vision Transformer (ViT-B/16)2026.0673.15——0.99
EWC*Model=Vision Transformer (ViT-B/16)2026.0670.94——1.54
DualPromptModel=Vision Transformer (ViT-B/16)2026.0664.8——8.74
L2PModel=Vision Transformer (ViT-B/16)2026.0661.08——9.23
LwFModel=Vision Transformer (ViT-B/16)2026.0660.94——16.24
EWCModel=Vision Transformer (ViT-B/16)2026.0650.59——27.62
CODA-PromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10—69.1974.18—
DualPromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10—68.3371.41—
DUCTBackbone=ViT-B/16-IN1K, Exemplars=02024.10—84.1485.1—
EASEBackbone=ViT-B/16-IN1K, Exemplars=02024.10—67.7864.96—
FinetuneBackbone=ViT-B/16-IN1K, Exemplars=02024.10—52.0850.11—
iCaRLBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10—68.4370.5—
L2PBackbone=ViT-B/16-IN1K, Exemplars=02024.10—67.3364.45—
MEMOBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10—60.8758.09—
RanPACBackbone=ViT-B/16-IN1K, Exemplars=02024.10—78.9280.48—
ReplayBackbone=ViT-B/16-IN1K, Exemplars=10 per class2024.10—66.9163.21—
S-iPromptBackbone=ViT-B/16-IN1K, Exemplars=02024.10—68.5172.76—
SimpleCILBackbone=ViT-B/16-IN1K, Exemplars=02024.10—60.863.4—