# The Pragmatic Persona: Discovering LLM Persona through Bridging Inference Source: [The Pragmatic Persona: Discovering LLM Persona through Bridging Inference]() Collection index: [llms.txt]() Data updated: 2026-06-13T05:00:05.486Z Authors: Jisoo Yang, Jongwon Ryu, Minuk Ma, Trung X. Pham, Junyeong Kim Published: 2026-04-27T06:01:47.000Z ## 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 ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Persona Discovery on Qwen3-1.7B Small Target](): Persona Discovery; Qwen3-1.7B Small Target; Similarity Score: 98. - [Persona Discovery on Llama 8B Small Target 3.1](): Persona Discovery; Llama 8B Small Target 3.1; Similarity Score: 0.97. - [Persona Discovery on Gemini Flash (Small Target) 2.5](): Persona Discovery; Gemini Flash (Small Target) 2.5; Similarity Score: 98. - [Persona Discovery on Qwen3-30B Large Target](): Persona Discovery; Qwen3-30B Large Target; Similarity Score: 99. - [Persona Discovery on Llama-3.1-70B Large Target](): Persona Discovery; Llama-3.1-70B Large Target; Similarity Score: 98. - [Persona Discovery on Qwen3-80B Large Target](): Persona Discovery; Qwen3-80B Large Target; Similarity Score: 99. ## Original sources and implementation [Original paper]() [PDF]()