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

Trajectory completion judgment on OGRBench Web

87.4Accuracy

OS-Themis

60.04867.14974.2581.351Jun 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
87.486.7
2026.06
87.487.2
2026.06
86.286.2
2026.06
85.386.7
2026.06
83.184.2
2026.06
82.683.6
2026.06
82.682.2
2026.06
82.581.8
2026.06
81.682.1
2026.06
81.680
2026.06
80.579.6
2026.06
79.581.2
2026.06
79.582.5
2026.06
7982.2
2026.06
77.975.3
2026.06
68.463
2026.06
66.861.8
2026.06
66.763.6
2026.06
65.360.7
2026.06
61.147.1