Dialogue Generation on Douban (test) benchmark leaderboard · SOTA2 Research

Dialogue Generation on Douban (test)

0.1398BLEU-1Ours

Evaluation Results

Method
Ours2022.05	0.1398	0.0516	0.0188	0.0074	0.4691	0.5091	0.6179
Ours (Hierarchical Contrastive Learning Framework)2022.05	0.1398	0.0516	0.0188	0.0074	0.4691	0.5091	0.6179
T5-CLAPS2022.05	0.1162	0.0413	0.0165	0.0062	0.4667	0.5071	0.6159
T5-CLAPS2022.05	0.1162	0.0413	0.0165	0.0062	0.4667	0.5071	0.6159
T52022.05	0.1007	0.0379	0.014	0.0057	0.4659	0.5056	0.6147
T52022.05	0.1007	0.0379	0.014	0.0057	0.4659	0.5056	0.6147
Dialo-gpt2022.05	0.0953	0.0363	0.0136	0.0057	0.4633	0.4997	0.6108
DialoGPT2022.05	0.0953	0.0363	0.0136	0.0057	0.4633	0.4997	0.6108
Seq2Seq-DU2022.05	0.0887	0.0333	0.0123	0.005	0.4591	0.4972	0.6083
Seq2Seq-DU2022.05	0.0887	0.0333	0.0123	0.005	0.4591	0.4972	0.6083
BERT-GEN2022.05	0.0823	0.0314	0.0119	0.005	0.463	0.5001	0.6095
BERT-GEN2022.05	0.0823	0.0314	0.0119	0.005	0.463	0.5001	0.6095
CVAE2022.05	0.064	0.0259	0.0102	0.0047	0.4473	0.4814	0.6006
CVAE2022.05	0.064	0.0259	0.0102	0.0047	0.4473	0.4814	0.6006
Group-wise2022.05	0.0581	0.0239	0.0101	0.0053	0.4489	0.4903	0.6002
Group-wise2022.05	0.0581	0.0239	0.0101	0.0053	0.4489	0.4903	0.6002
Seq2Seq2022.05	0.0542	0.0218	0.0086	0.0039	0.4388	0.4802	0.596
Seq2Seq2022.05	0.0542	0.0218	0.0086	0.0039	0.4388	0.4802	0.596
Transformer2022.05	0.0531	0.021	0.0081	0.0036	0.4401	0.4807	0.5989
Transformer2022.05	0.0531	0.021	0.0081	0.0036	0.4401	0.4807	0.5989