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

MSRA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Named Entity RecognitionMSRA (test)
F1 Score96.24
63
3D hand pose estimationMSRA
Mean Error (mm)7.13
32
Named Entity RecognitionMSRA
F1 Score96.01
29
3D Hand Pose EstimationMSRA (test)
3D Error (mm)5.186
23
ClusteringMSRA25
ARI56.3
20
Chinese Word SegmentationMSRA (test)
F1 Score98.67
17
Named Entity RecognitionMSRA standard (test)
F1 Score95
16
Unsupervised Feature SelectionMSRA
NMI86.48
14
ClusteringMSRA
Accuracy92.7
14
Hand Pose EstimationMSRA (leave-one-subject-out)
Mean Error (mm)7.2
12
Saliency DetectionMSRA5K
Max F-measure90.9
9
Saliency DetectionMSRA
F-measure86.66
8
Saliency DetectionMSRA
AUC95.83
8
Nonstationary backward transferMSRA25
BWT-ARI0.055
7
Incremental ClusteringMSRA25
Average Incremental ARI56
7
ClassificationMSRA
Accuracy93.71
7
ClusteringMSRA (test)
ACC71.43
7
Named Entity RecognitionMSRA
Inference Speed (sentences/s)5.95
7
SegmentationMSRA10k Degrade (test)
IoU86.09
5
Named Entity RecognitionChinese MSRA
F1 Score93.48
5
Hand Action RecognitionMSRA
Video Accuracy98.56
3
Showing 21 of 21 rows