List your company

Machine Translation on MT (test)

58.24Average Win RateALARM
45.000848.437951.87555.3121Mar 11, 2024

Evaluation results

MethodLinks
ALARMEvaluator=Grammar Error Rate2024.0358.2457.857.5359.37—
ALARMEvaluator=Holistic Reward2024.0352.1251.752.9351.74—
ALARMEvaluator=gpt-3.5-turbo2024.0351.7351.0352.1552.02—
Holistic RewardEvaluator=gpt-3.5-turbo2024.0350.43—51.0351.2748.97
Holistic RewardEvaluator=Holistic Reward2024.0350—51.5250.1848.3
Weighted SumEvaluator=Holistic Reward2024.0349.849.8251.31—48.26
Weighted SumEvaluator=gpt-3.5-turbo2024.0349.2448.7351.02—47.98
Holistic RewardEvaluator=Grammar Error Rate2024.0349.19—51.953.4742.2
Grammar RewardEvaluator=gpt-3.5-turbo2024.0348.648.97—48.9847.85
Grammar RewardEvaluator=Holistic Reward2024.0348.0848.47—48.6947.07
Grammar RewardEvaluator=Grammar Error Rate2024.0347.0748.1—50.6342.47
Weighted SumEvaluator=Grammar Error Rate2024.0345.5146.5449.37—40.63