List your company

Open-domain Dialogue Evaluation on Free run Mechanical Turk 1 (initial data collection run)

0.534Overall ScoreBi-Encoder Transformer
-0.27408-0.064290.14550.35529Mar 11, 2022

Evaluation results

MethodLinks
Bi-Encoder Transformerpersona=false, n=7982022.030.5340.5640.6020.7110.8630.964-0.0380.069
Poly-Encoder Transformerpersona=false, n=7982022.030.4190.4740.4810.6140.8750.994-0.431-0.075
Bi-Encoder Transformerpersona=true, n=7072022.030.3180.3990.3720.4430.8210.404-0.330.116
Key-Value Memory Networkpersona=false, n=7912022.030.2620.4910.3790.0280.636-0.066-0.3160.68
Key-Value Memory Networkpersona=true, n=7142022.030.1890.4090.3730.1590.672-0.114-0.5210.349
Poly-Encoder Transformerpersona=true, n=7072022.030.1730.230.1970.3690.6730.32-0.395-0.187
Sequence to Sequencepersona=false, n=7072022.03-0.087-0.19-0.2080.1660.3110.401-0.637-0.449
Sequence to Sequencepersona=true, n=7982022.03-0.201-0.308-0.2340.0920.3120.025-0.625-0.669
LSTM-based Modelpersona=true, n=7632022.03-0.217-0.181-0.201-0.1960.38-0.455-0.605-0.264
LSTM-based Modelpersona=false, n=7422022.03-0.243-0.165-0.16-0.1420.329-0.407-0.745-0.411