Salvato in:
Dettagli Bibliografici
Autori principali: Chu, KuanChao, Chen, Yi-Pei, Nakayama, Hideki
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2407.09897
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910562519089152
author Chu, KuanChao
Chen, Yi-Pei
Nakayama, Hideki
author_facet Chu, KuanChao
Chen, Yi-Pei
Nakayama, Hideki
contents This paper investigates the quality of multi-agent dialogues in simulations powered by Large Language Models (LLMs). Analyzing dialogues and memory over multiple sessions revealed significant issues such as repetition, inconsistency, and hallucination, exacerbated by the propagation of erroneous information. To combat these challenges, we propose a novel Screening, Diagnosis, and Regeneration (SDR) framework that detects and corrects utterance errors through a comprehensive process involving immediate issue identification, evidence gathering from past dialogues, and LLM analysis for utterance revision. By incorporating our SDR framework to Generative Agents (Park et al., 2023), we enhance the diversity, consistency, and factualness of the generated dialogues. This work presents a pioneering approach to enhancing dialogue quality in multi-agent simulations, establishing a new standard for future research in the field.
format Preprint
id arxiv_https___arxiv_org_abs_2407_09897
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Cohesive Conversations: Enhancing Authenticity in Multi-Agent Simulated Dialogues
Chu, KuanChao
Chen, Yi-Pei
Nakayama, Hideki
Computation and Language
This paper investigates the quality of multi-agent dialogues in simulations powered by Large Language Models (LLMs). Analyzing dialogues and memory over multiple sessions revealed significant issues such as repetition, inconsistency, and hallucination, exacerbated by the propagation of erroneous information. To combat these challenges, we propose a novel Screening, Diagnosis, and Regeneration (SDR) framework that detects and corrects utterance errors through a comprehensive process involving immediate issue identification, evidence gathering from past dialogues, and LLM analysis for utterance revision. By incorporating our SDR framework to Generative Agents (Park et al., 2023), we enhance the diversity, consistency, and factualness of the generated dialogues. This work presents a pioneering approach to enhancing dialogue quality in multi-agent simulations, establishing a new standard for future research in the field.
title Cohesive Conversations: Enhancing Authenticity in Multi-Agent Simulated Dialogues
topic Computation and Language
url https://arxiv.org/abs/2407.09897