[2604.04261] APPA: Adaptive Preference Pluralistic Alignment for Fair Federated RLHF of LLMs

Computer Science > Machine Learning arXiv:2604.04261 (cs) [Submitted on 5 Apr 2026] Title:APPA: Adaptive Preference Pluralistic Alignment for Fair Federated RLHF of LLMs Authors:Mahmoud Srewa, Tianyu Zhao, Salma Elmalaki View a PDF of the paper titled APPA: Adaptive Preference Pluralistic Alignment for Fair Federated RLHF of LLMs, by Mahmoud Srewa and 2 other authors View PDF HTML (experimental) Abstract:Aligning large language models (LLMs) with diverse human preferences requires pluralistic alignment, where a single model must respect the values of multiple distinct groups simultaneously. In federated reinforcement learning from human feedback (FedRLHF), these groups align a shared policy without centralizing preference data, which makes fair reward aggregation essential. Existing aggregation methods exhibit clear trade offs: average based aggregation systematically under aligns worst performing groups, while min aggregation prioritizes worst group performance at the cost of overall alignment. We propose APPA, an Adaptive Preference Pluralistic Alignment framework that dynamically reweights group level rewards based on historical alignment rewards. Our approach prioritizes under aligned groups without degrading well aligned ones, while requiring no access to raw preference data. Integrated into a proximal policy optimization (PPO) based FedRLHF pipeline and evaluated on GLOBALQA and OQA across three model families (Gemma 2 2B, Llama 3.2 3B, Qwen3 0.6B), APPA achieves str...