Fine-grained Visual Pattern Recognition on MMVP-VLM benchmark leaderboard · SOTA2 Research

Fine-grained Visual Pattern Recognition on MMVP-VLM

60Orientation ScoreUniME

Evaluation Results

Method	Links
UniME2026.02		60	43.3	73.3	63.3	56.7	76.7	60	43.3	50	58.5
AGFF-EMBED2026.02		60	43.3	73.3	73.3	46.7	83.3	66.7	50	56.7	61.5
LLaVE2026.02		53.3	40	66.7	63.3	43.3	73.3	53.3	30	60	53.7
UME-R12026.02		53.3	46.7	66.7	60	43.3	83.3	53.3	40	60	56.3
QQMM2026.02		53.3	43.3	73.3	46.7	36.7	83.3	50	43.3	56.7	54.1
UniME-V22026.02		46.7	46.7	73.3	56.7	50	73.3	63.3	43.3	60	57
DIVA2026.02		26.7	20	33.3	13.3	13.3	46.7	26.7	6.7	40	25.2
DFN2026.02		20	26.7	73.3	26.7	26.7	66.7	46.7	20	53.3	39.9
CLIP-IN2026.02		20	26.7	73.3	26.7	33.3	66.7	46.7	26.7	53.3	41.5
SigLIP22026.02		13.3	20	60	26.7	6.7	80	53.3	20	40	35.6
CLIP2026.02		0	20	40	20	6.7	20	33.3	6.7	40	20