Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Reasoning Evaluation on DeepfakeJudge Reason 1.0 (test)

9BLEU-1

Qwen-3-VL-30B-Instruct

0.682.8457.16Feb 23, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
9333662018623.31
8333462017602.79
2026.02
7223141815592.36
6223061812602.82
2026.02
5223051717603.17
2026.02
5222751715562.44
2026.02
5222951812602.7
2026.02
5222631515572.33
2026.02
4113041716603.59
2026.02
3112631515593.21
2026.02
3112741615603.43
2026.02
2112531413582.81
2026.02
11115185352.83
2026.02
1112011112562.93
2026.02
111161109532.51
2026.02
1112431615582.32