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

Multimodal Evaluation Consistency on MLLM-as-a-Judge

39.6CO Score

GPT-4o

12.35219.42626.533.574Jun 3, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.06
39.645.234.146.44656.440.857.358.930.526.256.942.134.2-
2025.06
38.940.743.456.849.559.63460.750.122.7-0.5402037.3-
2025.06
38.940.743.456.849.559.63460.750.122.7-0.5402037.3-
2025.06
38.24510.331.635.637.817.942.132.224.630.126.939.527.2-
2025.06
37368.643.145.746.820.7534426.514.735.227.142.2-
2025.06
33.346.314.645.247.455.939.654.548.827.325.933.440.337.4-
2025.06
32.841.125.154.151.358.341.462.147.124.323.650.728.739.8-
2025.06
32.841.125.154.151.358.341.462.147.124.323.650.728.739.8-
2025.06
30.13322.244.344.352.832.857.440.122.46.536.229.331.2-
2025.06
29.335.122.34446.643.42955.337.92517.243.130.633.6-
2025.06
29.335.122.34446.643.42955.337.92517.243.130.633.6-
2025.06
28.934.210.617.218.221.420.922.422.622.88.917.436.815.7-
2025.06
27.512.49.720.131.631.38.92635.125.631.527.226.337.2-
2025.06
26.4394.626.235.832.719.52941.514.435.926.744.425.3-
2025.06
26.437.328.5574951.243.256.443.327.2-0.139.624.829.3-
2025.06
26.437.328.5574951.243.256.443.327.2-0.139.624.829.3-
2025.06
25.934.315.137.928.538.417.63234.120.123.115.639.232.3-
2025.06
2537.819.634.440.54422.639.535.922.92236.932.421.8-
2025.06
2537.819.634.440.54422.639.535.922.92236.932.421.8-
2025.06
22.42.46.318.99.726.5-13.527.422.78.1326.124.926.2-
2025.06
20.619.413.621.114.828.24.930.329.616.934.213.542.339.1-
2025.06
13.410.312.215.413.721.319.713.514.216.118.111.936.315.8-