List your company

Pairwise Trajectory Preference Evaluation on DriveCritic (test)

82.83Accuracy (ACCTPS)DriveJudge-8B
48.686857.550966.41575.2791Jun 15, 2026

Evaluation results

MethodLinks
DriveJudge-8BFinetuning=true2026.0682.83—
DriveJudge-2BFinetuning=true2026.0680.63—
DriveCriticFinetuning=true2026.0676.33—
ADE/FDEFinetuning=false2026.0670.53—
Supervised ClassifierFinetuning=true2026.0664.8—
EPDMSFinetuning=false2026.0656.5—
Gemini3.1Finetuning=false2026.0655.45—
GPT-5Finetuning=false2026.0655.2—
RandomFinetuning=false2026.0650—
DriveCriticFine-tuning=✓2025.10—76
EPDMSFine-tuning=✗2025.10—41.4
GPT-5Fine-tuning=✗, zero-shot=true2025.10—55.2
OpenAI-o3Fine-tuning=✗, zero-shot=true2025.10—53.3
Qwen2.5-VL-7BFine-tuning=✗, zero-shot=true2025.10—48
Supervised Pairwise ClassifierFine-tuning=✓, Backbone=ResNet-1012025.10—64.8