HyperAIHyperAI초신경

Command Palette

Search for a command to run...

플랫폼

Personalized And Emotional Conversation On

평가 지표

Average Embedding
BLEU
Distinct-1
Distinct-2
Greedy Embedding
PPL
bertscore

평가 결과

이 벤치마크에서 각 모델의 성능 결과

Paper Title
GPT-{per+emo}0.56170.14030.06020.33880.502617.700.5719CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
GPT-{da}0.56100.13720.06050.33890.501717.720.5703CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
GPT-{per+emo+da}0.56080.13820.06010.34040501217.800.5722CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
GPT-{per}0.56060.13720.05920.33630.500918.080.5715CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
GPT-{emo}0.55880.13420.06140.34300.499617.480.5709CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
{emo+da}-GPT w/o emo0.55640.12520.04510.27460.496422.840.5666CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
{emo+da}-GPT w/o da0.55560.12720.04730.27900.496222.090.5669CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
{emo+da}-GPT0.55520.13040.04760.27850.496221.600.5674CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
GPT0.55090.11710.04820.27380.492220.070.5629CPED: A Large-Scale Chinese Personalized and Emotional Dialogue Dataset for Conversational AI
0 of 9 row(s) selected.