Enregistré dans:
Détails bibliographiques
Auteurs principaux: Ahn, Jaewoo, Lee, Taehyun, Lim, Junyoung, Kim, Jin-Hwa, Yun, Sangdoo, Lee, Hwaran, Kim, Gunhee
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2405.18027
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911891151912960
author Ahn, Jaewoo
Lee, Taehyun
Lim, Junyoung
Kim, Jin-Hwa
Yun, Sangdoo
Lee, Hwaran
Kim, Gunhee
author_facet Ahn, Jaewoo
Lee, Taehyun
Lim, Junyoung
Kim, Jin-Hwa
Yun, Sangdoo
Lee, Hwaran
Kim, Gunhee
contents While Large Language Models (LLMs) can serve as agents to simulate human behaviors (i.e., role-playing agents), we emphasize the importance of point-in-time role-playing. This situates characters at specific moments in the narrative progression for three main reasons: (i) enhancing users' narrative immersion, (ii) avoiding spoilers, and (iii) fostering engagement in fandom role-playing. To accurately represent characters at specific time points, agents must avoid character hallucination, where they display knowledge that contradicts their characters' identities and historical timelines. We introduce TimeChara, a new benchmark designed to evaluate point-in-time character hallucination in role-playing LLMs. Comprising 10,895 instances generated through an automated pipeline, this benchmark reveals significant hallucination issues in current state-of-the-art LLMs (e.g., GPT-4o). To counter this challenge, we propose Narrative-Experts, a method that decomposes the reasoning steps and utilizes narrative experts to reduce point-in-time character hallucinations effectively. Still, our findings with TimeChara highlight the ongoing challenges of point-in-time character hallucination, calling for further study.
format Preprint
id arxiv_https___arxiv_org_abs_2405_18027
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TimeChara: Evaluating Point-in-Time Character Hallucination of Role-Playing Large Language Models
Ahn, Jaewoo
Lee, Taehyun
Lim, Junyoung
Kim, Jin-Hwa
Yun, Sangdoo
Lee, Hwaran
Kim, Gunhee
Computation and Language
While Large Language Models (LLMs) can serve as agents to simulate human behaviors (i.e., role-playing agents), we emphasize the importance of point-in-time role-playing. This situates characters at specific moments in the narrative progression for three main reasons: (i) enhancing users' narrative immersion, (ii) avoiding spoilers, and (iii) fostering engagement in fandom role-playing. To accurately represent characters at specific time points, agents must avoid character hallucination, where they display knowledge that contradicts their characters' identities and historical timelines. We introduce TimeChara, a new benchmark designed to evaluate point-in-time character hallucination in role-playing LLMs. Comprising 10,895 instances generated through an automated pipeline, this benchmark reveals significant hallucination issues in current state-of-the-art LLMs (e.g., GPT-4o). To counter this challenge, we propose Narrative-Experts, a method that decomposes the reasoning steps and utilizes narrative experts to reduce point-in-time character hallucinations effectively. Still, our findings with TimeChara highlight the ongoing challenges of point-in-time character hallucination, calling for further study.
title TimeChara: Evaluating Point-in-Time Character Hallucination of Role-Playing Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2405.18027