MoCoRP: Modeling Consistent Relations between Persona and Response for Persona-based Dialogue

2025-12-08

AI Summary

Existing approaches inadequately model persona–response consistency in role-playing dialogue generation. Method: This paper proposes a novel paradigm that explicitly models the logical relationship between persona descriptions and generated responses. It introduces a natural language inference (NLI) expert model to automatically identify entailment, contradiction, or neutrality between them, thereby constructing a transferable consistency relation framework. Subsequently, relation-aware fine-tuning and alignment optimization inject structured logical signals into both BART and large language models during response generation. Results: The method achieves significant improvements on ConvAI2 and MPChat: +4.2% in persona consistency (human evaluation), +2.1 in BLEU, and +3.6 in F1 score—demonstrating gains in both automatic metrics and human assessment. This work establishes an interpretable, transferable pathway for persona consistency modeling.

Technology Category

• Natural Language Processing: Generation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Knowledge Representation and Reasoning: Common-Sense Reasoning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

As dialogue systems become increasingly important across various domains, a key challenge in persona-based dialogue is generating engaging and context-specific interactions while ensuring the model acts with a coherent personality. However, existing persona-based dialogue datasets lack explicit relations between persona sentences and responses, which makes it difficult for models to effectively capture persona information. To address these issues, we propose MoCoRP (Modeling Consistent Relations between Persona and Response), a framework that incorporates explicit relations into language models. MoCoRP leverages an NLI expert to explicitly extract the NLI relations between persona sentences and responses, enabling the model to effectively incorporate appropriate persona information from the context into its responses. We applied this framework to pre-trained models like BART and further extended it to modern large language models (LLMs) through alignment tuning. Experimental results on the public datasets ConvAI2 and MPChat demonstrate that MoCoRP outperforms existing baselines, achieving superior persona consistency and engaging, context-aware dialogue generation. Furthermore, our model not only excels in quantitative metrics but also shows significant improvements in qualitative aspects. These results highlight the effectiveness of explicitly modeling persona-response relations in persona-based dialogue. The source codes of MoCoRP are available at https://github.com/DMCB-GIST/MoCoRP.

Problem

• Generates engaging persona-based dialogues with coherent personality
• Addresses lack of explicit persona-response relations in datasets
• Enhances persona consistency and context-aware dialogue generation

Innovation

• Explicitly extracts NLI relations between persona and response
• Incorporates persona information into responses via NLI expert
• Extends framework to large language models through alignment tuning