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

RoboRefIt

Benchmarks

Task NameDataset NameSOTA ResultTrend
AffordanceRoboRefIt
Score82.3
12
Grounding ReasoningRoboRefIt (test)
Accuracy89.3
11
Robotic AffordanceRoboRefit Bench
Score87.15
9
PointingRoboRefit
Accuracy85.58
8
Object ReferencingRoboRefIt (test)
Accuracy56.7
8
2D TaskRoboRefit
Accuracy88.15
7
2D GroundingRoboRefit
Box-Hit88.15
7
Referring Expression SegmentationRoboRefIt (testB)
Accuracy74.4
5
Referring Expression SegmentationRoboRefIt (testA)
Accuracy85.5
5
Object ReferenceRoboRefIt
Metric-
0
Showing 10 of 10 rows