Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Driving explanation classification on BDD-X H1 & H2
Loading...
91.45
Accuracy (%)
RACE
79.1572
82.3486
85.54
88.7314
Feb 2, 2026
Accuracy (%)
F1 (macro)
Cohen's Kappa
Fleiss' Kappa
Agreement Level Description
Updated 3mo ago
Evaluation Results
Method
Method
Links
Accuracy (%)
F1 (macro)
Cohen's Kappa
Fleiss' Kappa
Agreement Level Description
RACE
Prompting Strategy=CoT-SC
2026.02
91.45
88
0.91
0.83
-
GPT-4o
Prompting Strategy=CoT
2026.02
84.22
82
0.82
0.72
-
GPT-4o-mini
Prompting Strategy=CoT
2026.02
82.13
70
0.79
0.76
-
GPT-3.5-Turbo
Prompting Strategy=CoT
2026.02
79.63
66
0.76
0.74
-
Feedback
Search any
task
Search any
task