EpiPersona: Persona Projection and Episode Coupling for Pluralistic Preference Modeling

TL;DR

EpiPersona introduces a novel framework for separating stable personal traits from episode-specific factors in preference modeling, improving adaptability of LLMs across diverse and shifting user preferences.

Contribution

It proposes explicit persona-episode coupling via low-dimensional persona projection and shared codes, enhancing preference prediction in varied scenarios.

Findings

Outperforms baselines in preference prediction tasks.

Achieves significant gains in episodic-shift scenarios.

Remains effective with sparse preference data.

Abstract

Pluralistic alignment is essential for adapting large language models (LLMs) to the diverse preferences of individuals and minority groups. However, existing approaches often mix stable personal traits with episode-specific factors, limiting their ability to generalize across episodes. To address this challenge, we introduce EpiPersona, a framework for explicit persona-episode coupling. EpiPersona first projects noisy preference feedback into a low-dimensional persona space, where similar personas are aggregated into shared discrete codes. This process separates enduring personal characteristics from situational signals without relying on predefined preference dimensions. The inferred persona representation is then coupled with the current episode, enabling episode-aware preference prediction. Extensive experiments show that EpiPersona consistently outperforms the baselines. It achieves notable performance gains in hard episodic-shift scenarios, while remaining effective with sparse preference data.