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

Multi-modal Chain-of-Thought Reasoning on MME CoT

64.2F1 Score

Kimi k1.5

29.56838.55947.5556.541Oct 24, 2025
Updated 11d ago

Evaluation Results

MethodLinks
2025.10
64.29249.31.42.9082.292.272.2
2025.10
6485.451.22.1-15.19692100
2025.10
56.277.344.2-2.1-6.52.496.592.9100
2025.10
47.764.837.73.97.8-0.159.781.837.5
2025.10
46.864.236.91.63.8-0.666.487.245.7
2025.10
4360.433.40.60.30.994.789.3100
2025.10
40.657.331.5-2.4-2.4-2.455.471.938.8
39.857.330.5-3.5-4.8-2.292.885.7100
2025.10
36.357.326.6-0.20.3-0.695.490.8100
2025.10
36.25327.4-7-7.5-6.551.176.326
2025.10
30.950.922.2-3.4-3.8-391.583100