List your company

Multimodal Understanding on MMMU out-of-distribution

56.55AccuracyCalibRL
54.594855.102455.6156.1176Feb 22, 2026

Evaluation results

MethodLinks
CalibRL2026.0256.55
RL-PLUS2026.0255.88
GRPO2026.0255.44
LUFFY2026.0255.22
SFT+GRPO2026.0254.67
DAPO2026.0254.67