OccuReward: LLM-Guided Occupant-Centric Reward Shaping for Demographic Equity in Grid-Interactive Buildings

ArXi:2605.28168v1 Announce Type: new Large language models (LLMs) have nstrated promising capability in generating reward functions for deep reinforcement learning (DRL)-based building energy management. However, their potential to exhibit or exacerbate disparities in occupant comfort across heterogeneous graphic populations remains unexplored. We present OccuReward, a framework investigating how LLM-mediated reward design affects graphic equity. Our contribution is three-fold: the