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/POMP

POMP

Reported on 15 benchmarks across 8 tasks · 1 paper · 6 SOTA

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

Computer Vision6 results

Open Vocabulary Semantic SegmentationonCOCO-Stuff-171
HIoU· 2023-04-10
39.1
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Open Vocabulary Semantic SegmentationonPascalVOC-20
hIoU· 2023-04-10
84.4
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Object DetectiononLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Open Vocabulary Object DetectiononLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Open Vocabulary Semantic SegmentationonPascalVOC-20
mIoU· 2023-04-10
89.4
best: 97.9 (UMG-CLIP-L/14)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Open Vocabulary Semantic SegmentationonADE20K-150
mIoU
20.7
best: 38.2 (Mask-Adapter)

Natural Language Processing5 results

Prompt EngineeringonImageNet-R
Top-1 accuracy %· 2023-04-10
77.9
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Prompt EngineeringonImageNet-21k
Accuracy· 2023-04-10
25.3
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Prompt EngineeringonImageNet-S
Top-1 accuracy %· 2023-04-10
49.8
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Prompt EngineeringonImageNet-A
Top-1 accuracy %· 2023-04-10
51.6
SOTA
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
Prompt EngineeringonImageNet V2
Top-1 accuracy %· uses extra data
63.8
best: 65.31 (HPT++)

Methodology4 results

3DonLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
2D ClassificationonLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
2D Object DetectiononLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704
16konLVIS v1.0
AP novel-LVIS base training· 2023-04-10
25.2
best: 43.4 (LaMI-DETR)
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition arXiv:2304.04704