The Pragmatic Persona: Discovering LLM Persona through Bridging Inference

2026-04-27

AI Summary

This study addresses the limitations of existing approaches to personality detection in large language models (LLMs), which predominantly rely on surface-level lexical or stylistic features and fail to capture the deeper discourse structures underlying consistent personality expression. To overcome this, the work proposes a novel framework that integrates cognitive discourse theory with bridging inference to model personality through structured semantic relationships. By leveraging bridging inference to uncover implicit conceptual links in dialogue, the method constructs knowledge graphs that enable personality modeling at the level of discourse coherence. Evaluated across multiple LLMs and reasoning backbones, the approach significantly outperforms frequency- and style-based baselines, demonstrating superior performance in both semantic coherence and personality identification stability, thereby transcending the constraints of conventional surface-level analysis.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large Language Models (LLMs) reveal inherent and distinctive personas through dialogue. However, most existing persona discovery approaches rely on surface-level lexical or stylistic cues, treating dialogue as a flat sequence of tokens and failing to capture the deeper discourse-level structures that sustain persona consistency. To address this limitation, we propose a novel analytical framework that interprets LLM dialogue through bridging inference -- implicit conceptual relations that connect utterances via shared world knowledge and discourse coherence. By modeling these relations as structured knowledge graphs, our approach captures latent semantic links that govern how LLMs organize meaning across turns, enabling persona discovery at the level of discourse coherence rather than surface realizations. Experimental results across multiple reasoning backbones and target LLMs, ranging from small-scale models to 80B-parameter systems, demonstrate that bridging-inference graphs yield significantly stronger semantic coherence and more stable persona identification than frequency or style-based baselines. These results show that persona traits are consistently encoded in the structural organization of discourse rather than isolated lexical patterns. This work presents a systematic framework for probing, extracting, and visualizing latent LLM personas through the lens of Cognitive Discourse Theory, bridging computational linguistics, cognitive semantics, and persona reasoning in large language models. Codes are available at https://github.com/JiSoo-Yang/Persona_Bridging.git

Problem

• persona discovery
• discourse coherence
• bridging inference
• large language models
• semantic structure

Innovation

• bridging inference
• persona discovery
• discourse coherence
• knowledge graph
• large language models