Overview
This theme centers on human-centric AI and human-AI interaction, with a focus on personalizing and steering large language models toward individual users. Work spans mechanistic frameworks for interpretable personalization and contextual-bandit-based optimization of user profiles for retrieval-augmented LLM personalization.
Motivation
Human-centric AI aims to make models adapt to the needs, preferences, and contexts of individual users rather than serving a single average user. This theme focuses on personalizing and steering large language models in interpretable and controllable ways, treating personalization as both a mechanistic and an optimization problem.
Research Directions
The theme works on LLM personalization and steering, combining mechanistic interpretability with retrieval-augmented and bandit-based personalization. Two ACL 2026 Main papers anchor the current work: a mechanistic framework that identifies “preference heads” for interpretable personalization, and a contextual-bandit approach for optimizing user profiles in retrieval-augmented LLM personalization. See the publications list above for details.
Related Publications
-
Preference Heads in Large Language Models: A Mechanistic Framework for Interpretable Personalization
Weixu Zhang, Ye Yuan, Changjiang Han, Yuxing Tian, Zipeng Sun, Linfeng Du, Jikun Kang, Hong Kang, Xue Liu, Haolun Wu
ACL 2026 (Main) · 2026
-
Optimizing User Profiles via Contextual Bandits for Retrieval-Augmented LLM Personalization
Linfeng Du, Ye Yuan, Zichen Zhao, Fuyuan Lyu, Emiliano Penaloza, Xiuying Chen, Zipeng Sun, Jikun Kang, Laurent Charlin, Xue Liu, Haolun Wu
ACL 2026 (Main) · 2026
Impact Holders
Impact holders and user communities will be added as the project scope becomes clearer.