List your company

Date Understanding on FLORES-200 10-languages

74.8Performance (kaz_Cyrl)English Pivot Thought
11.98428.29244.660.908Nov 2, 2024

Evaluation results

MethodLinks
English Pivot ThoughtModel=GPT-4o, EP=true, CT=true2024.1174.85863.66473.252.469.640.45663.661.56
DIPModel=GPT-4o, EP=true, CT=true2024.1172.466.873.271.67666.477.67075.675.672.52
English PivotingModel=GPT-4o, EP=true, CT=false2024.1149.642.845.244.849.64847.243.242.843.245.64
DIP w/ EP w/o CTModel=GPT-4o, EP=true, CT=false2024.1149.644.846.84850.448.845.645.24446.847
DIPBackbone=Mixtral2024.1148.841.651.25251.248.445.649.645.257.249.08
DIP w/o EP w/o CTBackbone=Mixtral, English pivoting (EP)=false, chain-of-thought (CT)=false2024.114841.643.647.648.443.24451.639.241.640.64
Non-insertion PromptingModel=GPT-4o, EP=false, CT=false2024.114839.244.443.643.642.443.24641.643.643.56
DIP w/o EP w/o CTModel=GPT-4o, EP=false, CT=false2024.11483844.445.647.640.441.638.439.244.842.8
Standard PromptingModel=GPT-4o, EP=false, CT=false2024.1147.247.253.654.461.647.255.645.246.44650.44
DIP w/ EP w/o CTBackbone=Mixtral, English pivoting (EP)=true, chain-of-thought (CT)=false2024.1139.634.436.837.642.441.6344032.43635.78
Standard PromptingBackbone=Mixtral2024.1122.431.624.829.634.832.418.828.825.621.630.92
English Pivot ThoughtBackbone=Mixtral, English pivoting (EP)=true, chain-of-thought (CT)=true2024.1121.62429.625.225.62618.829.62421.625.01
Non-insertion PromptingBackbone=Mixtral2024.1114.417.21819.220.82613.220.811.214.418.52
English PivotingBackbone=Mixtral, English pivoting (EP)=true2024.1114.418.815.619.616.8189.223.61213.614.68