List your company

ResearchFormerly WizWand

MLLM-as-a-Judge

Benchmarks

TaskDatasetSOTA resultTrendResults

MLLM-as-a-Judge

75.78Accuracy

32

Multimodal Evaluation Consistency

MLLM-as-a-Judge

39.6CO Score

22

Large Multimodal Model Evaluation

MLLM-as-a-Judge v1.0 (test)

49Overall Score

16

Reward Modeling

MLLM-as-a-Judge (MaaJ)

72.18Accuracy

13

Human Consistency Evaluation

MLLM-as-a-Judge

30.3CO Consistency Score

11

Pointwise Scoring

MLLM-as-a-Judge in-domain v1.0 (test)

80.2ImageDC Score

9