ChatHaruhi: Reviving Anime Character in Reality via Large Language Model
Authors:
Cheng Li,
Ziang Leng,
Chenxi Yan,
Junyi Shen,
Hao Wang,
Weishi MI,
Yaying Fei,
Xiaoyang Feng,
Song Yan,
HaoSheng Wang,
Linkang Zhan,
Yaokai Jia,
Pingyu Wu,
Haozhen Sun
Abstract:
Role-playing chatbots built on large language models have drawn interest, but better techniques are needed to enable mimicking specific fictional characters. We propose an algorithm that controls language models via an improved prompt and memories of the character extracted from scripts. We construct ChatHaruhi, a dataset covering 32 Chinese / English TV / anime characters with over 54k simulated…
▽ More
Role-playing chatbots built on large language models have drawn interest, but better techniques are needed to enable mimicking specific fictional characters. We propose an algorithm that controls language models via an improved prompt and memories of the character extracted from scripts. We construct ChatHaruhi, a dataset covering 32 Chinese / English TV / anime characters with over 54k simulated dialogues. Both automatic and human evaluations show our approach improves role-playing ability over baselines. Code and data are available at https://github.com/LC1332/Chat-Haruhi-Suzumiya .
△ Less
Submitted 18 August, 2023;
originally announced August 2023.
