Long Tail Learning On Imagenet Lt

المقاييس

Top-1 Accuracy

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

اسم النموذج	Top-1 Accuracy	Paper Title	Repository
BatchFormer(ResNet-50, RIDE)	55.7	BatchFormer: Learning to Explore Sample Relationships for Robust Representation Learning	-
Online Feature Augmentation	35.3	Feature Space Augmentation for Long-Tailed Data	-
BALMS	41.8	Balanced Meta-Softmax for Long-Tailed Visual Recognition	-
VL-LTR (ViT-B-16)	77.2	VL-LTR: Learning Class-wise Visual-Linguistic Representation for Long-Tailed Visual Recognition	-
APA (SE-ResNet-50)	57.9	Adaptive Parametric Activation	-
DeiT-LT	59.1	DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets	-
ProCo (ResNeXt50)	58.0	Probabilistic Contrastive Learning for Long-Tailed Visual Recognition	-
ProCo (ResNet50)	60.2	Probabilistic Contrastive Learning for Long-Tailed Visual Recognition	-
cRT + SSP	51.3	Rethinking the Value of Labels for Improving Class-Imbalanced Learning	-
KCL	51.5	Exploring Balanced Feature Spaces for Representation Learning	-
BALLAD(ResNet-101)	70.5	A Simple Long-Tailed Recognition Baseline via Vision-Language Model	-
BatchFormer(ResNet-50, PaCo)	57.4	BatchFormer: Learning to Explore Sample Relationships for Robust Representation Learning	-
NCL(ResNet-50)	57.4	Nested Collaborative Learning for Long-Tailed Visual Recognition	-
OPeN (ResNeXt-50)	55.1	Pure Noise to the Rescue of Insufficient Data: Improving Imbalanced Classification by Training on Random Noise Images	-
TSC(ResNet-50)	52.4	Targeted Supervised Contrastive Learning for Long-Tailed Recognition	-
smDRAGON	42.0	From Generalized zero-shot learning to long-tail with class descriptors	-
RIDE (ResNeXt-50)	56.4	Long-tailed Recognition by Routing Diverse Distribution-Aware Experts	-
DRO-LT	53.5	Distributional Robustness Loss for Long-tail Learning	-
LDAM + DRW + SAM	53.1	Escaping Saddle Points for Effective Generalization on Class-Imbalanced Data	-
LIFT (ViT-B/16)	78.3	Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts	-

0 of 69 row(s) selected.