Saved in:
Bibliographic Details
Main Authors: Lotfi, Ehsan, De Bruyn, Maxime, Buhmann, Jeska, Daelemans, Walter
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.07363
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909073195139072
author Lotfi, Ehsan
De Bruyn, Maxime
Buhmann, Jeska
Daelemans, Walter
author_facet Lotfi, Ehsan
De Bruyn, Maxime
Buhmann, Jeska
Daelemans, Walter
contents The new wave of Large Language Models (LLM) has offered an efficient tool to curate sizeable conversational datasets. So far studies have mainly focused on task-oriented or generic open-domain dialogs, and have not fully explored the ability of LLMs in following complicated prompts. In this work, we focus on personalization, and employ LLMs to curate a dataset which is difficult and costly to crowd-source: PersonalityChat is a synthetic conversational dataset based upon the popular PersonaChat dataset, but conditioned on both personas and (Big-5) personality traits. Evaluating models fine-tuned on this dataset, we show that the personality trait labels can be used for trait-based personalization of generative dialogue models. We also perform a head-to-head comparison between PersonalityChat and PersonaChat, and show that training on the distilled dataset results in more fluent and coherent dialog agents in the small-model regime.
format Preprint
id arxiv_https___arxiv_org_abs_2401_07363
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PersonalityChat: Conversation Distillation for Personalized Dialog Modeling with Facts and Traits
Lotfi, Ehsan
De Bruyn, Maxime
Buhmann, Jeska
Daelemans, Walter
Computation and Language
The new wave of Large Language Models (LLM) has offered an efficient tool to curate sizeable conversational datasets. So far studies have mainly focused on task-oriented or generic open-domain dialogs, and have not fully explored the ability of LLMs in following complicated prompts. In this work, we focus on personalization, and employ LLMs to curate a dataset which is difficult and costly to crowd-source: PersonalityChat is a synthetic conversational dataset based upon the popular PersonaChat dataset, but conditioned on both personas and (Big-5) personality traits. Evaluating models fine-tuned on this dataset, we show that the personality trait labels can be used for trait-based personalization of generative dialogue models. We also perform a head-to-head comparison between PersonalityChat and PersonaChat, and show that training on the distilled dataset results in more fluent and coherent dialog agents in the small-model regime.
title PersonalityChat: Conversation Distillation for Personalized Dialog Modeling with Facts and Traits
topic Computation and Language
url https://arxiv.org/abs/2401.07363