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

Multimodal Sequencing on RecipeQA Golden (test)

92.12Accuracy

Human Performance

28.763245.211661.6678.1084Oct 16, 2021
Updated 5mo ago

Evaluation Results

MethodLinks
2021.10
92.1283.134.824.530.950.45
2021.10
88.9278.564.764.410.930.7
2021.10
82.6684.614.10.881.1
2021.10
80.464.54.544.020.861.29
2021.10
79.2604.573.930.851.29
2021.10
78.2604.563.910.851.44
2021.10
77574.493.810.841.48
2021.10
76.2584.493.850.831.58
2021.10
74524.453.680.831.64
2021.10
47.2163.682.40.524.12
2021.10
38.483.392.020.355.44
2021.10
31.253.272.070.276.1