Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas
About
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.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Future Interaction Prediction | Srv (test) | Hit Rate @ 1031.39 | 7 | |
| Future Interaction Prediction | MerRec (test) | Hit Rate@1058.65 | 7 | |
| Future Interaction Prediction | AOL (test) | Hit Rate @1040.18 | 7 | |
| Persona Induction | Srv (test) | Coherence0.455 | 7 | |
| Persona Induction | MerRec (test) | Coherence0.674 | 7 | |
| Persona Induction | AOL (test) | Coherence71.8 | 7 |