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

Driving Reasoning on BDD-X Hard

15.6BLEU-4

CPO++

11.4412.5213.614.68Apr 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
15.645.864.2
2026.04
13.640.552.7
2026.04
13.542.962.4
2026.04
1343.560.4
2026.04
12.342.157.3
2026.04
11.64249.8