List your company

Machine Translation on Flores-101 (val test)

46.8CHRFLSL-NAS
45.13645.5684646.432May 4, 2023

Evaluation results

MethodLinks
LSL-NASdense pre-training=true, Total Parameters (|θ|)=441M, Effective Parameters (|θeff|)=186M2023.0546.827.5
Separate Decoder Baselinehidden dim=768, Total Parameters (|θ|)=509M, Effective Parameters (|θeff|)=296M2023.0546.527.3
LSL-NASTotal Parameters (|θ|)=441M, Effective Parameters (|θeff|)=186M2023.0546.427.2
LSLSRC layers={1, 2}, TGT layers={15, 16}, Total Parameters (|θ|)=413M, Effective Parameters (|θeff|)=186M2023.0546.327
Separate Decoder Baselinehidden dim=704, Total Parameters (|θ|)=453M, Effective Parameters (|θeff|)=268M2023.0546.226.9
LSLSRC layers={1, 2, 3}, TGT layers={14, 15, 16}, Total Parameters (|θ|)=470M, Effective Parameters (|θeff|)=186M2023.0546.227
Language Adapters ENC (SRC+TGT)bottleneck dimension=256, Total Parameters (|θ|)=342M, Effective Parameters (|θeff|)=228M2023.0546.126.7
LSLSRC layers={3, 4}, TGT layers=0, Total Parameters (|θ|)=356M, Effective Parameters (|θeff|)=186M2023.0546.126.7
Language Adapters ENC (SRC+TGT)bottleneck dimension=512, Total Parameters (|θ|)=384M, Effective Parameters (|θeff|)=270M2023.054626.7
Separate Decoder Baselinehidden dim=640, Total Parameters (|θ|)=399M, Effective Parameters (|θeff|)=240M2023.0545.926.6
Language Adapters ENCbottleneck dimension=128, Total Parameters (|θ|)=321M, Effective Parameters (|θeff|)=207M2023.0545.826.4
LSLSRC layers=Ø, TGT layers={13, 14, 15}, Total Parameters (|θ|)=385M, Effective Parameters (|θeff|)=186M2023.0545.826.5
Language Adapters ENCbottleneck dimension=256, Total Parameters (|θ|)=342M, Effective Parameters (|θeff|)=228M2023.0545.726.3
Language Adapters ENC (SRC+TGT)bottleneck dimension=128, Total Parameters (|θ|)=321M, Effective Parameters (|θeff|)=207M2023.0545.726.3
Language Adapters ENCbottleneck dimension=512, Total Parameters (|θ|)=384M, Effective Parameters (|θeff|)=270M2023.0545.626.3
Separate Decoder BaselineTotal Parameters (|θ|)=299M, Effective Parameters (|θeff|)=186M2023.0545.526
LSLSRC layers={13, 14, 15}, TGT layers={3, 4}, Total Parameters (|θ|)=441M, Effective Parameters (|θeff|)=186M2023.0545.225.7