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/DRAGON

DRAGON

Reported on 46 benchmarks across 6 tasks · 2 papers · 22 SOTA

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

Methodology30 results

Generalized Few-Shot ClassificationonCUB-LT
Long-Tailed Accuracy· 2020-04-05
67.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonAWA-LT
Long-Tailed Accuracy· 2020-04-05
94.1
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonSUN-LT
Long-Tailed Accuracy· 2020-04-05
40.4
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonCUB-LT
Long-Tailed Accuracy· 2020-04-05
67.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonAWA-LT
Long-Tailed Accuracy· 2020-04-05
94.1
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonSUN-LT
Long-Tailed Accuracy· 2020-04-05
40.4
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (10-shots)· 2020-04-05
81.9
best: 82.2 (MVCN)
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (20-shots)· 2020-04-05
83.3
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (1-shot)· 2020-04-05
41
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (10-shots)· 2020-04-05
48.2
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (2-shots)· 2020-04-05
43.8
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (5-shots)· 2020-04-05
46.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonCUB-LT
Long-Tailed Accuracy· 2020-04-05
67.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAWA-LT
Long-Tailed Accuracy· 2020-04-05
94.1
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN-LT
Long-Tailed Accuracy· 2020-04-05
40.4
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonCUB-LT
Per-Class Accuracy· 2020-04-05
57.8
best: 60.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonAWA-LT
Per-Class Accuracy· 2020-04-05
74.1
best: 76.2 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonSUN-LT
Per-Class Accuracy· 2020-04-05
34.8
best: 36.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot ClassificationonImageNet-LT-d
Per-Class Accuracy· 2020-04-05
51.2
best: 53.5 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonCUB-LT
Per-Class Accuracy· 2020-04-05
57.8
best: 60.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonAWA-LT
Per-Class Accuracy· 2020-04-05
74.1
best: 76.2 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonSUN-LT
Per-Class Accuracy· 2020-04-05
34.8
best: 36.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Long-tail LearningonImageNet-LT-d
Per-Class Accuracy· 2020-04-05
51.2
best: 53.5 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (1-shot)· 2020-04-05
67.1
best: 69.9 (MVCN)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (2-shots)· 2020-04-05
69.1
best: 76.4 (MVCN)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (5-shots)· 2020-04-05
76.7
best: 81.2 (MVCN)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonCUB-LT
Per-Class Accuracy· 2020-04-05
57.8
best: 60.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonAWA-LT
Per-Class Accuracy· 2020-04-05
74.1
best: 76.2 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonSUN-LT
Per-Class Accuracy· 2020-04-05
34.8
best: 36.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Generalized Few-Shot LearningonImageNet-LT-d
Per-Class Accuracy· 2020-04-05
51.2
best: 53.5 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235

Computer Vision14 results

Image ClassificationonCUB-LT
Long-Tailed Accuracy· 2020-04-05
67.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonAWA-LT
Long-Tailed Accuracy· 2020-04-05
94.1
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonSUN-LT
Long-Tailed Accuracy· 2020-04-05
40.4
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonCUB-LT
Long-Tailed Accuracy· 2020-04-05
67.7
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonAWA-LT
Long-Tailed Accuracy· 2020-04-05
94.1
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonSUN-LT
Long-Tailed Accuracy· 2020-04-05
40.4
SOTA
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonCUB-LT
Per-Class Accuracy· 2020-04-05
57.8
best: 60.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonAWA-LT
Per-Class Accuracy· 2020-04-05
74.1
best: 76.2 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonSUN-LT
Per-Class Accuracy· 2020-04-05
34.8
best: 36.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Image ClassificationonImageNet-LT-d
Per-Class Accuracy· 2020-04-05
51.2
best: 53.5 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonCUB-LT
Per-Class Accuracy· 2020-04-05
57.8
best: 60.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonAWA-LT
Per-Class Accuracy· 2020-04-05
74.1
best: 76.2 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonSUN-LT
Per-Class Accuracy· 2020-04-05
34.8
best: 36.1 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235
Few-Shot Image ClassificationonImageNet-LT-d
Per-Class Accuracy· 2020-04-05
51.2
best: 53.5 (DRAGON + Bal'Loss)
From Generalized zero-shot learning to long-tail with class descriptors arXiv:2004.02235

Natural Language Processing2 results

Common Sense ReasoningonRiddleSense
Accuracy (%)· 2022-10-17
71.3
SOTA
Deep Bidirectional Language-Knowledge Graph Pretraining arXiv:2210.09338
Common Sense ReasoningonCommonsenseQA
Accuracy· 2022-10-17
78.2
best: 92.54 (GPT-4o (HPT))
Deep Bidirectional Language-Knowledge Graph Pretraining arXiv:2210.09338