arXiv AI October 27, 2026 By Yujie Zhang, Weikang Yuan, Zhuoren Jiang, Pengwei Yan

EpiPersona: Persona Projection and Episode Coupling for Pluralistic Preference Modeling

EpiPersona is a framework that explicitly couples persona and episode information to improve pluralistic preference modeling for large language models. It projects noisy preference feedback into a low‑dimensional persona space, aggregates similar personas into discrete codewords, and then couples the inferred persona representation with the current episode for episode‑aware preference prediction. Experiments demonstrate that EpiPersona outperforms baselines, especially in hard episodic‑shift scenarios and with sparse preference data.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jul 28

Group Preference Collapse in Personalized Multimodal Large Language Models

arXiv:2607. 22603v1 Announce Type: new Abstract: Personalized multimodal large language models (MLLMs) aim to generate user-specific responses, but existing methods mainly rely on profile-level information and overlook diverse user preferences.

By Fan Lyu, Wenqi Zhang, Joost van de Weijer

llms fine-tuning multimodal

arXiv Machine Learning

Jun 30

Synthetic Interaction Data for Scalable Personalization in Large Language Models

arXiv:2602. 12394v2 Announce Type: replace Abstract: Personalized prompting offers large opportunities for deploying large language models (LLMs) to diverse users, yet existing prompt optimization methods primarily focus on task-level optimization while largely overlooking user-specific preferences and latent constraints of individual users.

By Yuchen Ma, Yue Huang, Wenjie Wang, Xiaonan Luo, Xiangliang Zhang, Stefan Feuerriegel

llms agents reinforcement-learning benchmarks safety

arXiv AI

Sep 1

Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment

The paper introduces Personalized Group Relative Policy Optimization (P‑GRPO), a new alignment framework for large language models that separates advantage estimation from immediate batch statistics. By normalizing advantages using preference‑group‑specific reward histories instead of the concurrent generation group, P‑GRPO maintains contrastive signals for distinct user preferences. Experiments across various tasks show that P‑GRPO converges faster and yields higher rewards than standard GRPO, improving alignment with heterogeneous human preferences while preserving general capabilities.

By Jialu Wang, Heinrich Peters, Asad A. Butt, Navid Hashemi, Alireza Hashemi, Pouya M. Ghari, Joseph Hoover, James Rae, Morteza Dehghani

llms reinforcement-learning safety

arXiv Computation and Language

Sep 2

FaST: Feature-aware Sampling and Tuning for Personalized Preference Alignment with Limited Data

arXiv:2508.04698v2 Announce Type: replace Abstract: LLM-powered conversational assistants are often deployed in a one-size-fits-all manner, which fails to accommodate individual user preferences. Rec...

By Thibaut Thonet, Germ\'an Kruszewski, Jos Rozen, Pierre Erbacher, Marc Dymetman

llms benchmarks safety

arXiv Computation and Language

Sep 22

Pretrained Persona Mixture Models and Tandem Models for Human Simulation

arXiv:2609.22607v1 Announce Type: new Abstract: We argue here that the current dominant practice in LLM human simulation: prompting instruction-tuned assistant language models to role-play personas,...

By Minwoo Kang, T\'ea Wright, Seun Eisape, Ayush Raj, Suhong Moon, Joseph Suh, Alane Suhr, David M. Chan, John Canny

llms

arXiv Machine Learning

Jul 30

Learning Dynamic User Personas from Implicit Interaction Streams via Iterative Refinement

arXiv:2607. 26473v1 Announce Type: new Abstract: Personalizing large language models (LLMs) to individual users is essential for improving user experience, yet existing approaches typically rely on explicit preference supervision such as pairwise comparisons or demographic attributes, limiting their applicability in natural interaction settings.

By Haifeng Wu

llms agents robotics