ReasonMatch-Bench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ReasonMatch-Bench Overall 90-sample high-divergence subset
94.7F1 Score
18
ReasonMatch-Bench Object (test)
45.6L1 Metric
10
ReasonMatch-Bench Outdoor (test)
90.9L1 Score
10
ReasonMatch-Bench Indoor (test)
25.5L1 Error
10
ReasonMatch-Bench Overall (test)
70.5F1 Score
10
ReasonMatch-Bench uCO3D high-divergence
62.1F1 Score
6
ReasonMatch-Bench DL3DV high-divergence
93.5F1 Score
6