Cohesive Conversations: Enhancing Authenticity in Multi-Agent Simulated Dialogues

TL;DR

This paper introduces a novel SDR framework to improve the authenticity, consistency, and factual accuracy of multi-agent dialogues generated by LLMs, addressing issues like repetition and hallucination.

Contribution

We propose a comprehensive SDR framework that detects, diagnoses, and regenerates dialogue errors, significantly improving dialogue quality in multi-agent simulations.

Findings

Enhanced dialogue diversity and consistency

Reduced hallucination and repetition

Improved factual accuracy in generated dialogues

Abstract

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.