List your company
ResearchFormerly WizWand

Saliency Detection on PASCAL-Context (test)

85.78maxFUV-M3TL
64.896870.318475.7481.1616Nov 22, 2023Apr 3, 2024Aug 15, 2024Dec 27, 2024May 10, 2025Sep 21, 2025Feb 2, 2026
Updated 8mo ago

Evaluation results

MethodLinks
UV-M3TLBackbone=ViT-large2026.0285.78——
MTMamba++Backbone=Swin-Large2026.0285.56——
TaskDiffusionBackbone=ViT-large2026.0284.94——
MLOREBackbone=ViT-large2024.0384.9——
MLoREBackbone=ViT-large2026.0284.9——
TaskExpertBackbone=ViT-large2024.0384.87——
TaskExpertBackbone=ViT-large2026.0284.87——
TSP-TransformerBackbone=ViT-large2026.0284.86——
TaskPrompterBackbone=ViT-large2024.0384.83——
TaskPrompterBackbone=ViT-large2026.0284.83——
InvPTBackbone=ViT-large2024.0384.81——
InvPTBackbone=ViT-large2026.0284.81——
MTI-NetBackbone=HRNet182024.0384.78——
MTI-NetBackbone=HRNet182026.0284.78——
MTI-NetBackbone=ViT-large, reproduced=true2024.0384.75——
MTI-Net (Reproduced)Backbone=ViT-large2026.0284.75——
InvPT++Backbone=ViT-large2026.0284.74——
MLOREFLOPs (G)=72, #Param (M)=442024.0384.7——
ATRCBackbone=HRNet182026.0284.7——
TaskExpertFLOPs (G)=204, #Param (M)=552024.0384.68——
MTMambaBackbone=Swin-Large2026.0284.14——
ATRC-ASPPBackbone=HRNet182026.0283.91——
MQTransformerBackbone=not specified2026.0283.58——
FedAMPClient Setup=Multi-Task2023.1183.350.27—
LocalClient Setup=Multi-Task2023.1183.230—
FEDHCA2Client Setup=Multi-Task2023.1183.180.75—
MaT-FLClient Setup=Multi-Task2023.1182.84-0.16—
FedPerClient Setup=Multi-Task2023.1182.5-0.16—
ATRCBackbone=HRNet182024.0382.29——
ATRC-BMTASBackbone=HRNet182026.0282.29——
DittoClient Setup=Multi-Task2023.1181.74-4.67—
FedProxClient Setup=Multi-Task2023.1181.29-8.12—
ATRCBackbone=ViT-large, reproduced=true2024.0381.2——
ATRC (Reproduced)Backbone=ViT-large2026.0281.2——
FedAvgClient Setup=Multi-Task2023.1181.08-7.56—
LocalBackbone=Swin-T, Client Setup=Single-Task2023.1180.91——
FedAMPBackbone=Swin-T, Client Setup=Single-Task2023.1180.79——
FEDHCA^2Backbone=Swin-T, Client Setup=Single-Task2023.1180.71——
GDBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0679.6—67.99
Aligned-MTLBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0679.57—67.94
PCGradBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0679.38—67.86
MaT-FLBackbone=Swin-T, Client Setup=Single-Task2023.1179.26——
OursBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0679.26—68.3
PAD-NetBackbone=ViT-large, reproduced=true2024.0379.21——
PAD-Net (Reproduced)Backbone=ViT-large2026.0279.21——
CAGradBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0679.12—67.46
IndependentBackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0678.98—67.05
MGDABackbone=HRNet-18, Task weights=manually tuned, Experimental seeds=average of 32024.0678.98—67.32
FedPerBackbone=Swin-T, Client Setup=Single-Task2023.1178.85——
DittoBackbone=Swin-T, Client Setup=Single-Task2023.1177.99——
FedAvgBackbone=Swin-T, Client Setup=Single-Task2023.1177.56——
FedProxBackbone=Swin-T, Client Setup=Single-Task2023.1177.26——
Mod-SquadFLOPs (G)=420, #Param (M)=522024.0366.9——
M³ViTFLOPs (G)=420, #Param (M)=422024.0366.3——
PAD-NetBackbone=HRNet182024.0365.8——
PAD-NetBackbone=HRNet182026.0265.8——
ASTMTBackbone=not specified2026.0265.7——
AdapterBackbone=Swin-Tiny, Trainable Parameters (M)=11.24, Single Inference For All Tasks=false2024.03——61.28
BitfitBackbone=Swin-Tiny, Trainable Parameters (M)=2.85, Single Inference For All Tasks=false2024.03——60.64
CompactorBackbone=Swin-Tiny, Trainable Parameters (M)=2.78, Single Inference For All Tasks=false2024.03——60.08
Compactor++Backbone=Swin-Tiny, Trainable Parameters (M)=2.66, Single Inference For All Tasks=false2024.03——59.47
HyperFormerBackbone=Swin-Tiny, Trainable Parameters (M)=72.77, Single Inference For All Tasks=false2024.03——65.54
LoRABackbone=Swin-Tiny, Trainable Parameters (M)=2.87, Single Inference For All Tasks=false2024.03——61.9
MTL - Full Fine TuningBackbone=Swin-Tiny, Trainable Parameters (M)=30.06, Single Inference For All Tasks=true2024.03——65.21
MTL - Tuning Decoders OnlyBackbone=Swin-Tiny, Trainable Parameters (M)=1.94, Single Inference For All Tasks=true2024.03——57.46
MTLORABackbone=Swin-Tiny, Rank (r)=16, Trainable Parameters (M)=4.95, Single Inference For All Tasks=true2024.03——64.48
MTLORABackbone=Swin-Tiny, Rank (r)=32, Trainable Parameters (M)=6.08, Single Inference For All Tasks=true2024.03——64.9
MTLORABackbone=Swin-Tiny, Rank (r)=64, Trainable Parameters (M)=8.34, Single Inference For All Tasks=true2024.03——65.4
MTLORA+Backbone=Swin-Tiny, Rank (r)=4, Trainable Parameters (M)=2.57, Single Inference For All Tasks=true2024.03——63.14
MTLORA+Backbone=Swin-Tiny, Rank (r)=8, Trainable Parameters (M)=3.15, Single Inference For All Tasks=true2024.03——63.57
MTLORA+Backbone=Swin-Tiny, Rank (r)=16, Trainable Parameters (M)=4.29, Single Inference For All Tasks=true2024.03——64.32
PolyhistorBackbone=Swin-Tiny, Trainable Parameters (M)=8.96, Single Inference For All Tasks=false2024.03——65.54
Single TaskBackbone=Swin-Tiny, Trainable Parameters (M)=112.62, Single Inference For All Tasks=false2024.03——62.35
VL-AdapterBackbone=Swin-Tiny, Trainable Parameters (M)=4.74, Single Inference For All Tasks=false2024.03——62.29
VPT-deepBackbone=Swin-Tiny, Trainable Parameters (M)=3.43, Single Inference For All Tasks=false2024.03——58.15
VPT-shallowBackbone=Swin-Tiny, Trainable Parameters (M)=2.57, Single Inference For All Tasks=false2024.03——58.31