Multi-turn Response Selection on Douban Conversation Corpus

63.2MAPTripleNet (Ensemble)

Evaluation results

Method
TripleNet (Ensemble)2019.09	63.2	67.8	51.5	31.3	49.4	83.2
SA-BERT2021.05	61.9	65.9	49.6	31.3	48.1	84.7
TripleNet (ELMo)2019.09	60.9	65	47	27.8	48.7	81.4
BERT + SPIDER2021.05	60.9	65	47.5	29.6	48.8	83.6
BERT2021.05	60.5	64.7	47.4	29.1	47.8	84.9
BERT + SPIDER2021.05	59.8	63.8	45.9	28.5	48.7	82.6
BERT2021.05	58.8	63.1	45.3	27.7	46.4	81.8
MSN2021.05	58.7	63.2	47	29.5	45.2	78.8
IoI2021.05	57.3	62.1	44.4	26.9	45.1	78.6
MRFN2021.05	57.1	61.7	44.8	27.6	43.5	78.3
TripleNet2019.09	56.4	61.8	44.7	26.8	42.6	77.8
DUA2019.09	55.1	59.9	42.1	24.3	42.1	78
DUA2021.05	55.1	59.9	42.1	24.3	42.1	78
DAM2019.09	55	60.1	42.7	25.4	41	75.7
DAM2021.05	55	60.1	42.7	25.4	41	75.7
SMN2019.09	52.9	56.9	39.7	23.3	39.6	72.4
SMN2021.05	52.9	56.9	39.7	23.3	39.6	72.4
Multi-View2019.09	50.5	54.3	34.2	20.2	35	72.9
MV-LSTM2019.09	49.8	53.8	34.8	20.2	35.1	71.6
Match-LSTM2019.09	49.8	53.8	34.8	20.2	34.8	71
DL2R2019.09	48.8	52.7	33	19.3	34.2	70.5
DualEncoder2019.09	48.5	52.7	32	18.7	34.3	72
IMN2019.01	0.576	0.618	0.441	0.268	0.458	0.796
IMN2019.01	0.57	0.615	0.433	0.262	0.452	0.789
U2U-IMN2019.11	0.564	0.611	0.429	0.259	0.43	0.791
DUA2018.06	0.551	0.599	0.421	0.243	0.421	0.78
DUA2019.11	0.551	0.599	0.421	0.243	0.421	0.78
DUA2019.01	0.551	0.599	0.421	0.243	0.421	0.78
DAM2019.11	0.55	0.601	0.427	0.254	0.41	0.757
DAM2019.01	0.55	0.601	0.427	0.254	0.41	0.757
SMNdynamic2016.12	0.529	0.569	0.397	0.233	0.396	0.724
SMN2018.06	0.529	0.569	0.397	0.233	0.396	0.724
SMN2019.11	0.529	0.569	0.397	0.233	0.396	0.724
SMN2019.01	0.529	0.569	0.397	0.233	0.396	0.724
SMNlast2016.12	0.526	0.571	0.393	0.236	0.387	0.729
SMNstatic2016.12	0.523	0.572	0.387	0.228	0.387	0.734
Multi-Channel2016.12	0.506	0.543	0.349	0.203	0.351	0.709
Multi-Channel2018.06	0.506	0.543	0.349	0.203	0.351	0.709
Multi-View2016.12	0.505	0.543	0.342	0.202	0.35	0.729
Multi-View2018.06	0.505	0.543	0.342	0.202	0.35	0.729
Multi-View2019.11	0.505	0.543	0.342	0.202	0.35	0.729
Match-LSTM2016.12	0.5	0.537	0.345	0.202	0.348	0.72
Match-LSTM2018.06	0.5	0.537	0.345	0.202	0.348	0.72
Match-LSTM2019.11	0.5	0.537	0.345	0.202	0.348	0.72
MV-LSTM2016.12	0.498	0.538	0.348	0.202	0.351	0.71
MV-LSTM2018.06	0.498	0.538	0.348	0.202	0.351	0.71
MV-LSTM2019.11	0.498	0.538	0.348	0.202	0.351	0.71
Attentive-LSTM2016.12	0.495	0.523	0.331	0.192	0.328	0.718
Attentive-LSTM2018.06	0.495	0.523	0.331	0.192	0.328	0.718
DL2R2016.12	0.488	0.527	0.33	0.193	0.342	0.705
DL2R2018.06	0.488	0.527	0.33	0.193	0.342	0.705
DL2R2019.11	0.488	0.527	0.33	0.193	0.342	0.705
LSTM2016.12	0.485	0.527	0.32	0.187	0.343	0.72
LSTM2018.06	0.485	0.537	0.32	0.187	0.343	0.72
LSTM2019.11	0.485	0.527	0.32	0.187	0.343	0.72
BiLSTM2016.12	0.479	0.514	0.313	0.184	0.33	0.716
BILSTM2018.06	0.479	0.514	0.313	0.184	0.33	0.716
Multi-Channelexp2016.12	0.476	0.515	0.317	0.179	0.335	0.691
Multi-Channel_exp2018.06	0.476	0.515	0.317	0.179	0.335	0.691
CNN2016.12	0.417	0.44	0.226	0.121	0.252	0.647
CNN2018.06	0.417	0.44	0.226	0.121	0.252	0.647
RNN2016.12	0.39	0.422	0.208	0.118	0.223	0.589
RNN2018.06	0.39	0.422	0.208	0.118	0.223	0.589
RNN2019.11	0.39	0.422	0.208	0.118	0.223	0.589
TF-IDF2016.12	0.331	0.359	0.18	0.096	0.172	0.405
TF-IDF2018.06	0.331	0.359	0.18	0.096	0.172	0.405
TF-IDF2019.11	0.331	0.359	0.18	0.096	0.172	0.405