Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/DeiT-LT

DeiT-LT

Reported on 25 benchmarks across 5 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology15 results

Generalized Few-Shot ClassificationonCIFAR-100-LT (ρ=50)
Error Rate· 2024-04-03
39.5
best: 9.8 (LIFT (ViT-B/16, ImageNet-21K pre-training))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot ClassificationonImageNet-LT
Top-1 Accuracy· 2024-04-03
59.1
best: 82.9 (LIFT (ViT-L/14))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot ClassificationonCIFAR-10-LT (ρ=50)
Error Rate· 2024-04-03
10.2
best: 8.44 (GLMC + SAM)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot ClassificationonCIFAR-100-LT (ρ=100)
Error Rate· 2024-04-03
44.4
best: 10.9 (LPT)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot ClassificationonCIFAR-10-LT (ρ=100)
Error Rate· 2024-04-03
12.5
best: 10.42 (GLMC+MaxNorm (ResNet-34, channel x4))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Long-tail LearningonCIFAR-100-LT (ρ=50)
Error Rate· 2024-04-03
39.5
best: 9.8 (LIFT (ViT-B/16, ImageNet-21K pre-training))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Long-tail LearningonImageNet-LT
Top-1 Accuracy· 2024-04-03
59.1
best: 82.9 (LIFT (ViT-L/14))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Long-tail LearningonCIFAR-10-LT (ρ=50)
Error Rate· 2024-04-03
10.2
best: 8.44 (GLMC + SAM)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Long-tail LearningonCIFAR-100-LT (ρ=100)
Error Rate· 2024-04-03
44.4
best: 10.9 (LPT)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Long-tail LearningonCIFAR-10-LT (ρ=100)
Error Rate· 2024-04-03
12.5
best: 10.42 (GLMC+MaxNorm (ResNet-34, channel x4))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot LearningonCIFAR-100-LT (ρ=50)
Error Rate· 2024-04-03
39.5
best: 9.8 (LIFT (ViT-B/16, ImageNet-21K pre-training))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot LearningonImageNet-LT
Top-1 Accuracy· 2024-04-03
59.1
best: 82.9 (LIFT (ViT-L/14))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot LearningonCIFAR-10-LT (ρ=50)
Error Rate· 2024-04-03
10.2
best: 8.44 (GLMC + SAM)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot LearningonCIFAR-100-LT (ρ=100)
Error Rate· 2024-04-03
44.4
best: 10.9 (LPT)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Generalized Few-Shot LearningonCIFAR-10-LT (ρ=100)
Error Rate· 2024-04-03
12.5
best: 10.42 (GLMC+MaxNorm (ResNet-34, channel x4))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900

Computer Vision10 results

Image ClassificationonCIFAR-100-LT (ρ=50)
Error Rate· 2024-04-03
39.5
best: 9.8 (LIFT (ViT-B/16, ImageNet-21K pre-training))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Image ClassificationonImageNet-LT
Top-1 Accuracy· 2024-04-03
59.1
best: 82.9 (LIFT (ViT-L/14))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Image ClassificationonCIFAR-10-LT (ρ=50)
Error Rate· 2024-04-03
10.2
best: 8.44 (GLMC + SAM)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Image ClassificationonCIFAR-100-LT (ρ=100)
Error Rate· 2024-04-03
44.4
best: 10.9 (LPT)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Image ClassificationonCIFAR-10-LT (ρ=100)
Error Rate· 2024-04-03
12.5
best: 10.42 (GLMC+MaxNorm (ResNet-34, channel x4))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Few-Shot Image ClassificationonCIFAR-100-LT (ρ=50)
Error Rate· 2024-04-03
39.5
best: 9.8 (LIFT (ViT-B/16, ImageNet-21K pre-training))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Few-Shot Image ClassificationonImageNet-LT
Top-1 Accuracy· 2024-04-03
59.1
best: 82.9 (LIFT (ViT-L/14))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Few-Shot Image ClassificationonCIFAR-10-LT (ρ=50)
Error Rate· 2024-04-03
10.2
best: 8.44 (GLMC + SAM)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Few-Shot Image ClassificationonCIFAR-100-LT (ρ=100)
Error Rate· 2024-04-03
44.4
best: 10.9 (LPT)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900
Few-Shot Image ClassificationonCIFAR-10-LT (ρ=100)
Error Rate· 2024-04-03
12.5
best: 10.42 (GLMC+MaxNorm (ResNet-34, channel x4))
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets arXiv:2404.02900