List your company
ResearchFormerly WizWand

Fine-grained classification on DTD

86.58AccuracyLoRA
20.352837.546454.7471.9336Jan 13, 2021Dec 5, 2021Oct 28, 2022Sep 20, 2023Aug 12, 2024Jul 5, 2025May 28, 2026
Updated 5mo ago

Evaluation results

MethodLinks
LoRAStudent=ViT-B, Teacher=ViT-L2026.0586.58
SLADStudent=ViT-B, Teacher=ViT-L2026.0586.4
LoRAStudent=ViT-S, Teacher=ViT-B2026.0584.71
SLADStudent=ViT-S, Teacher=ViT-B2026.0584.56
SLADStudent=ViT-S, Teacher=ViT-L2026.0584.24
LoRAStudent=ViT-S, Teacher=ViT-L2026.0584.18
Probing*Student=ViT-B, Teacher=ViT-L2026.0583.49
ProbingStudent=ViT-S, Teacher=ViT-L2026.0583.4
Probing*Student=ViT-S, Teacher=ViT-B2026.0581.93
Probing*Student=ViT-S, Teacher=ViT-L2026.0581.61
StableRepPre-training Dataset=CC12M, Pre-training Data Type=Syn, Pre-training Schedule=105 epochs, Backbone=ViT-B/162023.0680.1
StableRepPre-training Dataset=RedCaps, Pre-training Data Type=Syn, Pre-training Schedule=105 epochs, Backbone=ViT-B/162023.0680
CLIPPre-training Dataset=WIT-400M, Pre-training Data Type=Real, Backbone=ViT-B/162023.0679.2
StableRepPre-training Dataset=RedCaps, Pre-training Data Type=Syn, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0679
StableRepPre-training Dataset=CC12M, Pre-training Data Type=Syn, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0678.1
SimCLRPre-training Dataset=RedCaps, Pre-training Data Type=Real, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0677.9
CLIPPre-training Dataset=RedCaps, Pre-training Data Type=Syn, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0676
ResNet-50 (ReLabel-trained)Backbone=ResNet-50, Pre-training strategy=ReLabel2021.0175.74
ResNet-50 (Baseline)Backbone=ResNet-50, Pre-training strategy=Standard2021.0175.43
SimCLRPre-training Dataset=RedCaps, Pre-training Data Type=Syn, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0674.6
CLIPPre-training Dataset=RedCaps, Pre-training Data Type=Real, Pre-training Schedule=35 epochs, Backbone=ViT-B/162023.0673.9
ADAPTBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0855.73
ADAPTBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0851.89
DMNBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0850.41
DPEBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0850.18
BCABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0848.58
StatABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0848.5
TPSBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0848.4
TransCLIPBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Transductive2025.0847.8
DotaBackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0845.8
TDABackbone=CLIP-RN50, BP-free=true, Evaluation Protocol=Online2025.0843.74
C-TPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0842.2
TPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0840.84
DiffTPTBackbone=CLIP-RN50, BP-free=false, Evaluation Protocol=Online2025.0840.72
CLIPBackbone=CLIP-RN50, BP-free=N/A, Evaluation Protocol=Online2025.0840.37
xCLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.1033.2
CLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.1027.6
nCLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.1022.9