Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ADE

Benchmarks

Task NameDataset NameSOTA ResultTrend
Open-Vocabulary Semantic SegmentationADE-847
mIoU18.1
63
Semantic SegmentationADE-847
mIoU17.6
47
Semantic SegmentationADE
mIoU46.4
42
Joint Entity and Relation ExtractionADE
Entity F1 Score0.912
26
Promptable Image SegmentationADE (val)
mIoU74.9
24
Relation ExtractionADE
Relation Strict F183.9
20
Named Entity RecognitionADE (test)
F1 Score96.31
19
Image SegmentationADE
mIoU74.5
18
Open Vocabulary SegmentationADE CS
mIoU43.21
16
Open-Vocabulary Semantic SegmentationADE-150
mIoU33.5
15
Visual Question AnsweringADE
Accuracy (ADE)56.08
13
Relation ExtractionADE (test)
Macro F190
13
Promptable Image SegmentationADE
mIoU74.9
11
Relation ExtractionADE corpus
F1 Score84.45
11
SegmentationADE (out-domain)
PQ26.8
10
Interactive SegmentationADE
IoU (Point)66.4
9
Named Entity RecognitionADE exact match (10-fold cross val)
Macro-F191.3
7
ClassificationADE
F1 Score91.03
7
Semantic Image SynthesisADE-outd. (val)
FID48.6
7
Relation ExtractionADE 10-fold cross-validation
Macro F183.2
7
Open-vocabulary SegmentationADE-OV
mIoU23.7
6
Named Entity RecognitionADE
NER Accuracy91.5
6
Entity RecognitionADE 10-fold cross-validation
F1 Score0.8711
6
SegmentationADE150
mIoU24.7
5
Object DetectionADE-150 (val)
AP (Box)29.6
5
Showing 25 of 33 rows