UPCS: Unbiased Persona Construction for Dialogue Generation

TL;DR

The paper introduces UPCS, a framework for constructing unbiased persona profiles in dialogue systems, improving accuracy, diversity, and fairness in personalized narrative interactions.

Contribution

UPCS is the first framework to categorize persona descriptions into eight dimensions with bias mitigation strategies for dialogue systems.

Findings

UPCS outperforms existing methods in accuracy and diversity.

UPCS effectively reduces biases in persona profiles.

User satisfaction is significantly improved with UPCS.

Abstract

Narrative systems, such as dialogue and storytelling systems, often utilize persona profiles to enhance personalized interactions. Existing persona profiles frequently exhibit biases, posing risks to system integrity and fairness. To address this, we introduce the UPCS framework, which categorizes character descriptions into eight dimensions, including bias mitigation strategies. Experimental results demonstrate UPCS's superiority in accuracy, diversity, bias elimination, and user satisfaction, marking a significant advancement in persona construction for reliable narrative systems.