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

XLRS-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Visual Question AnsweringXLRS-Bench L-3 Capability (test)
OC61.7
33
Remote Sensing Image UnderstandingXLRS-Bench
Accuracy52.78
20
Remote Sensing Perception and ReasoningXLRS-Bench
Average Score (Avg.)53.1
19
Remote Sensing ReasoningXLRS-Bench
PASS@160.4
18
Remote Sensing Visual Question AnsweringXLRS-Bench
Average Score0.542
17
Visual Question AnsweringXLRS-Bench vqa
F1 Score21.6
10
Image CaptioningXLRS-Bench caption
GEval Score40.4
10
Scientific ReasoningXLRS-Bench Lite
Score51.12
9
Remote SensingXLRS-Bench
Score52.8
5
Showing 9 of 9 rows