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

TIGER

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multiple Instance Learning ClassificationTIGER
Accuracy96.9
20
Multiagent offline planningTiger SDec
Planning Value109.0925
18
Time predictionTIGeR 86k (test)
ToY Error48.86
16
POMDP AbstractionTiger full actions
Validation Error Rate0
10
Geo-localizationTIGeR 86k (test)
Recall@200km48.63
8
SegmentationTIGER (test)
Dice Score45.9
8
WSI-level segmentationTIGER
Dice72.6
8
RecommendationTIGER Tools Backbone (Period 3)
H@52.55
7
Geo-time Aware Image RetrievalTIGeR 86k (test)
Recall@13.51
5
SegmentationTIGER wsirois (internal held-out)
Dice Score59
5
SegmentationTIGER wsibulk (internal held-out)
Dice89.2
5
Compositional Image RetrievalTIGeR 86k (test)
Recall@117.84
4
POMDP PlanningTiger
PBVI Original Time (s)0.001
1
Multi-instance Learning ClassificationTIGER classical MIL (10-fold cross-val)
Metric-
0
Showing 14 of 14 rows