List your company

Human Preference Prediction on internal benchmark

82.2AccuracyQwen-7B
63.37668.26373.1578.037Apr 30, 2026

Evaluation results

MethodLinks
Qwen-7BConfiguration=GCPO2026.0482.2
Seed-1.5-VLConfiguration=T+V (Think+Verify)2026.0479.3
Seed-1.6-VLConfiguration=T+V (Think+Verify)2026.0477.2
Qwen-7BConfiguration=T+V (Think+Verify)2026.0475.4
Seed-1.5-VLConfiguration=T (Think)2026.0472.2
Qwen-3BConfiguration=GCPO2026.0472
Seed-1.6-VLConfiguration=T (Think)2026.0471.2
Qwen-7BConfiguration=V (Verify)2026.0470.9
Seed-1.6-VLConfiguration=V (Verify)2026.0469.4
Qwen-3BConfiguration=T+V (Think+Verify)2026.0469.3
Qwen-7BConfiguration=T (Think)2026.0468.9
Qwen-7B (VIESCORE)Configuration=V (Verify)2026.0468.3
Qwen-3BConfiguration=V (Verify)2026.0466.1
Qwen-3BConfiguration=T (Think)2026.0464.1