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

Abstract Visual Reasoning on VCog-Bench

42.4CVR Score

Ours (RL)

22.53627.69332.8538.007Jan 31, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
42.412.1
2026.01
40.811.3
2026.01
39.812.1
2026.01
38.212.3
2026.01
36.713
2026.01
34.512.5
2026.01
33.310.4
2026.01
32.310.8
2026.01
23.323.4