List your company
ResearchFormerly WizWand

Image Classification on CIFAR100 (Baseline Accuracy)

94.1AccuracyFranca
77.730481.980286.2390.4798Mar 25, 2024Aug 5, 2024Dec 16, 2024Apr 28, 2025Sep 8, 2025Jan 19, 2026Jun 1, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FrancaViT=L/14, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0594.1
DINOv2ViT=L/14, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0593.9
LoRA (r = 32)Backbone=DINOv2-ViT-B/14, Params (M)=4.722026.0493.7
RandLoRA (r = 10)Backbone=DINOv2-ViT-B/14, Params (M)=3.592026.0493.7
BaLoRABackbone=ViT-L/16, Pre-trained=ImageNet12K2026.0493.62
Full Fine-TuningBackbone=DINOv2-ViT-B/14, Params (M)=86.732026.0493.6
LoRABackbone=ViT-L/16, Pre-trained=ImageNet12K2026.0493.59
FFBackbone=ViT-Large, Method=FF, # Params=303.3M2026.0593.58
DoRABackbone=ViT-L/16, Pre-trained=ImageNet12K2026.0493.44
DreamNet-3MAE-lType=DreamNet, Input Size=224^2, #Params=209.90, FLOPs=149.42024.0993.4
DINOv2ViT=L/14, Pre-training Data=LVD-142M, Evaluation Protocol=Linear, Notes=distilled from DINOv2-G2026.0593.4
FourierFTBackbone=ViT-Large, Method=FourierFT, # Params=480K2026.0593.37
FourierFTBackbone=ViT-Large, Method=FourierFT, # Params=144K2026.0593.31
Full Fine-tuningBackbone=ViT-L/16, Pre-trained=ImageNet12K2026.0493.25
GiVA (r = 32, TopSV-B)Backbone=DINOv2-ViT-B/14, Params (M)=0.082026.0493.1
LoRA (r = 32)Backbone=CLIP-ViT-L/14, Params (M)=12.582026.0493.1
Uni-LoRABackbone=ViT-Large, Method=Uni-LoRA, # Params=144K2026.0593.08
VeRA (r = 256)Backbone=DINOv2-ViT-B/14, Params (M)=0.092026.0493
GiVA (r = 32, SecSV-B)Backbone=DINOv2-ViT-B/14, Params (M)=0.082026.0493
GiVA (r = 32, Rand-B)Backbone=DINOv2-ViT-B/14, Params (M)=0.082026.0493
RandLoRA (r = 10)Backbone=CLIP-ViT-L/14, Params (M)=12.782026.0493
Full Fine-TuningBackbone=CLIP-ViT-L/14, Params (M)=303.282026.0492.8
iBOTViT=L/16, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0592.8
GPartBackbone=ViT-Large, Method=GPart (Ours), # Params=144K2026.0592.59
VeRA (r = 256)Backbone=CLIP-ViT-L/14, Params (M)=0.232026.0492.4
GiVA (r = 32, TopSV-B)Backbone=CLIP-ViT-L/14, Params (M)=0.202026.0492.4
GiVA (r = 32, Rand-B)Backbone=CLIP-ViT-L/14, Params (M)=0.202026.0492.4
FFBackbone=ViT-Base, Method=FF, # Params=85.8M2026.0592.38
SleepNet-3ViT-lType=SleepNet, Input Size=224^2, #Params=13.40, FLOPs=371.52024.0992.3
DreamNet-3Type=DreamNet, Input Size=224^2, #Params=209.12, FLOPs=83.52024.0992.3
DreamNet-3MAE-bType=DreamNet, Input Size=224^2, #Params=209.12, FLOPs=83.52024.0992.3
GiVA (r = 32, SecSV-B)Backbone=CLIP-ViT-L/14, Params (M)=0.202026.0492.3
SleepNet-3Type=SleepNet, Input Size=224^2, #Params=10.07, FLOPs=60.62024.0992.2
SleepNet-3ViT-bType=SleepNet, Input Size=224^2, #Params=10.07, FLOPs=107.32024.0992.2
GPartBackbone=ViT-Base, Method=GPart (Ours), # Params=72K2026.0592.11
EfficientNetV2-LType=Conv only, Input Size=480^2, #Params=121.32, FLOPs=53.02024.0992.1
CvT-W24Type=Conv + Trans, Input Size=384^2, #Params=277.33, FLOPs=193.22024.0992.1
Uni-LoRABackbone=ViT-Base, Method=Uni-LoRA, # Params=72K2026.0592.1
TC-JEPAViT=L/16, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0591.6
FourierFTBackbone=ViT-Base, Method=FourierFT, # Params=239K2026.0591.45
MAElargeType=Transformer, Input Size=224^2, #Params=304.18, FLOPs=61.92024.0991.3
ViTl-ACNType=Augmentation +Trans, Input Size=384^2, #Params=490.11, FLOPs=41.92024.0991.2
TC-JEPAViT=L/16, Pre-training Data=CC27M, Evaluation Protocol=Linear2026.0591.2
SigLIP2ViT=L/16, Pre-training Data=WebLI, Evaluation Protocol=Linear2026.0591.2
FourierFTBackbone=ViT-Base, Method=FourierFT, # Params=72K2026.0591.2
MAEbaseType=Transformer, Input Size=224^2, #Params=86.43, FLOPs=17.62024.0991.1
SleepNet-4Type=SleepNet, Input Size=224^2, #Params=10.14, FLOPs=80.72024.0991.1
ViTlargeType=Transformer, Input Size=384^2, #Params=307.12, FLOPs=190.72024.0991
SleepNet-3MAE-lType=SleepNet, Input Size=224^2, #Params=10.09, FLOPs=126.52024.0991
Web-DINOViT=L/16, Pre-training Data=MC-2B, Evaluation Protocol=Linear2026.0590.7
SleepNet-3MAE-bType=SleepNet, Input Size=224^2, #Params=10.07, FLOPs=60.62024.0990.2
EfficientNet-B7Type=Conv only, Input Size=600^2, #Params=66.44, FLOPs=37.02024.0990.1
CvT-21Type=Conv + Trans, Input Size=384^2, #Params=32.41, FLOPs=24.92024.0990.1
MAEl-MASType=Augmentation +Trans, Input Size=224^2, #Params=551.42, FLOPs=29.92024.0990
SigLIPViT=L/16, Pre-training Data=WebLI, Evaluation Protocol=Linear2026.0589.8
CAPIViT=L/14, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0589.2
I-JEPAViT=L/16, Pre-training Data=IN-21k, Evaluation Protocol=Linear2026.0588.7
ViTbaseType=Transformer, Input Size=224^2, #Params=86.45, FLOPs=55.42024.0987.3
ResNet50Type=Conv only, Input Size=224^2, #Params=25.12, FLOPs=3.82024.0986.9
DreamNet-2Type=DreamNet, Input Size=224^2, #Params=208.31, FLOPs=52.32024.0985.1
DreamNet-4Type=DreamNet, Input Size=224^2, #Params=212.03, FLOPs=118.42024.0985.1
LPBackbone=ViT-Base, Method=LP, # Params=02026.0584.28
LPBackbone=ViT-Large, Method=LP, # Params=02026.0584.28
ResNet101Type=Conv only, Input Size=224^2, #Params=45.01, FLOPs=7.62024.0984.1
SAFformerParam=6.342026.0583.38
SleepNet-2Type=SleepNet, Input Size=224^2, #Params=10.01, FLOPs=40.52024.0983.1
Max-FormerParam=6.602026.0582.65
QB-LIFType=SNN Training, Architecture=ResNet-20, Timestep=42026.0482.25
STAA-SNN2026.0582.05
MSViTParam=7.592026.0581.98
QB-LIFType=SNN Training, Architecture=ResNet-20, Timestep=22026.0481.72
QB-LIFType=SNN Training, Architecture=ResNet-20, Timestep=12026.0481.36
Spikinghash-MParam (M)=6.24, Time Step=42025.0181.28
QKFormerParam (M)=6.74, Time Step=42025.0181.15
QKFormerParam=6.742026.0581.15
QKFormerParam (M)=6.74, Time Step (T)=42024.0381.15
Trasnformer (ANN)Param (M)=9.32, Time Step (T)=12024.0381.02
FACTSampling Ratio=10%, Backbone=ViT2026.0680.66
ResNet18Type=Conv only, Input Size=224^2, #Params=11.23, FLOPs=1.82024.0980.6
Burst+LIPoolingType=ANN2SNN, Architecture=ResNet-20, Timestep=2562026.0480.57
FSTA-SNNParam=11.302026.0580.42
Spikingformer-CMLParam (M)=9.32, Time Step=42025.0180.37
AXISArchitecture=ViT-B-322025.0880.13
CMLParam (M)=9.32, Time Step (T)=42024.0380.02
Spikinghash-SParam (M)=3.87, Time Step=42025.0179.51
KvLIFNetwork Architecture=ResNet-18, Timestep=62026.0379.3
SpikingResformer-TiParam (M)=10.79, Time Step=42025.0179.28
aTLASArchitecture=ViT-B-322025.0879.14
SpikingformerParam (M)=9.32, Time Step=42025.0179.09
L3FACTSampling Ratio=10%, Backbone=ViT2026.0678.78
RecDis-SNNType=SNN Training, Architecture=ResNet-19, Timestep=42026.0478.53
ILIFNetwork Architecture=ResNet-18, Timestep=62026.0378.51
DSRNetwork Architecture=PreAct-ResNet-18, Timestep=202026.0378.5
BLS-SeTaBackbone=ResNet18, Pruning Rate=30%2026.0478.5
SeTaBackbone=ResNet18, Pruning Rate=30%2026.0478.4
BLS-InfoBatchBackbone=ResNet18, Pruning Rate=30%2026.0478.4
Spike-driven TransformerParam (M)=9.32, Time Step=42025.0178.4
S-TransformerParam=10.282026.0578.4
S-TransformerParam (M)=10.28, Time Step (T)=42024.0378.4
CLIFNetwork Architecture=ResNet-18, Timestep=62026.0378.36