List your company
ResearchFormerly WizWand

Image Classification on Average across 10 datasets

74.4Average AccuracyVR (glass-box)
53.70459.07764.4569.823Dec 19, 2025Jan 5, 2026Jan 22, 2026Feb 8, 2026Feb 25, 2026Mar 14, 2026Apr 1, 2026
Updated 7mo ago

Evaluation results

MethodLinks
VR (glass-box)Backbone=CLIP ViT-B/16 (Service), Shot Setting=16-shot, #API (M)=16.82, Time (h)=9.82026.0474.4
AReS-MSBackbone=Service & Local selection, Shot Setting=16-shot, #API (M)=0.06, Time (h)=3.72026.0469.3
ManiPTZero-shot=true, Source=ImageNet2026.0268.04
CLIP-LoRA + ADKBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1267.4
2SFS + ADKBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1267.4
Rep-Adapter + ADKBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1267.1
CoPromptZero-shot=true, Source=ImageNet2026.0266.99
TAPZero-shot=true, Source=ImageNet2026.0266.56
TACZero-shot=true, Source=ImageNet2026.0266.53
MaPLeZero-shot=true, Source=ImageNet2026.0266.5
CoCoOpZero-shot=true, Source=ImageNet2026.0265.74
AReSBackbone=ViT-B/16 (Local), Shot Setting=16-shot, #API (M)=0.02, Time (h)=3.72026.0465.4
2SFSBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1265.1
CLIPBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1265
CLIP-LoRABackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1264.6
Rep-AdapterBackbone=ViT-B/16, Training Dataset=ImageNet, Evaluation Protocol=all-to-all2025.1264.3
CoOpZero-shot=true, Source=ImageNet2026.0263.88
BlackVIPBackbone=CLIP ViT-B/16 (Service), Shot Setting=16-shot, #API (M)=754.20, Time (h)=197.52026.0462.9
BARBackbone=CLIP ViT-B/16 (Service), Shot Setting=16-shot, #API (M)=612.84, Time (h)=185.62026.0461.4
Zero-shotBackbone=CLIP ViT-B/16 (Service), Shot Setting=16-shot, #API (M)=0.12, Time (h)=02026.0457.2
LLM-OptBackbone=CLIP ViT-B/16 (Service), Shot Setting=16-shot, #API (M)=5011.32, Time (h)=5.12026.0454.5