Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas

TL;DR

This paper introduces a hierarchical framework for inducing multiple, evidence-grounded, and truthful user personas from behavioral logs, improving coherence and trustworthiness over existing methods.

Contribution

The authors propose a novel hierarchical approach with an optimization-based training method to generate more coherent and trustworthy user personas from logs.

Findings

Induces more coherent and evidence-grounded personas.

Improves trustworthiness of generated personas.

Enhances future interaction prediction accuracy.

Abstract

Behavioral logs provide rich signals for user modeling, but are noisy and interleaved across diverse intents. Recent work uses LLMs to generate interpretable natural-language personas from user logs, yet evaluation often emphasizes downstream utility, providing limited assurance of persona quality itself. We propose a hierarchical framework that aggregates user actions into intent memories and induces multiple evidence-grounded personas by clustering and labeling these memories. We formulate persona induction as an optimization problem over persona quality-captured by cluster cohesion, persona-evidence alignment, and persona truthfulness-and train the persona model using a groupwise extension of Direct Preference Optimization (DPO). Experiments on a large-scale service log and two public datasets show that our method induces more coherent, evidence-grounded, and trustworthy personas, while also improving future interaction prediction.