HyperAIHyperAI超神经

Command Palette

Search for a command to run...

算力平台

Out Of Distribution Detection On Imagenet 1K 8

评估指标

AUROC
FPR95

评测结果

各个模型在此基准测试上的表现结果

Paper Title
DOE76.2680.94Out-of-distribution Detection with Implicit Outlier Transformation
RP+GradNorm-73.53Detecting Out-of-distribution Data through In-distribution Class Prior
KNN (ResNet-50)80.1069.53Out-of-Distribution Detection with Deep Nearest Neighbors
DML-57.40Decoupling MaxLogit for Out-of-Distribution Detection
SHE-54.19Out-of-Distribution Detection based on In-Distribution Data Patterns Memorization with Modern Hopfield Energy
ODIN+UMAP (ResNet-50)86.9249.69Unleashing Mask: Explore the Intrinsic Out-of-Distribution Detection Capability
KNN (ResNet-50 SupCon)88.4048.91Out-of-Distribution Detection with Deep Nearest Neighbors
ReAct (ResNet-50)88.1647.69ReAct: Out-of-distribution Detection With Rectified Activations
GradNorm (ResNetv2-101)-46.48On the Importance of Gradients for Detecting Distributional Shifts in the Wild
NPOS90.4443.77Non-Parametric Outlier Synthesis
MOS (BiT-S-R101x1)92.0140.63MOS: Towards Scaling Out-of-distribution Detection for Large Semantic Space
MCM (CLIP-B)92.5737.59Delving into Out-of-Distribution Detection with Vision-Language Representations
DICE (ResNet-50)90.8335.15DICE: Leveraging Sparsification for Out-of-Distribution Detection
RankFeat (ResNetv2-101)94.0729.27RankFeat: Rank-1 Feature Removal for Out-of-distribution Detection
MCM (CLIP-L)94.1429.00Delving into Out-of-Distribution Detection with Vision-Language Representations
ASH-S (ResNet-50)94.0227.98Extremely Simple Activation Shaping for Out-of-Distribution Detection
DICE + ReAct (ResNet-50)93.9425.45DICE: Leveraging Sparsification for Out-of-Distribution Detection
SCALE (ResNet50)95.0223.27Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancement
NNGuide (RegNet)94.4321.58Nearest Neighbor Guidance for Out-of-Distribution Detection
LINe (ResNet50)95.2619.48LINe: Out-of-Distribution Detection by Leveraging Important Neurons
0 of 21 row(s) selected.