← all papers · overview

Mitigating Reward Hacking In RLHF Via Bayesian Non-negative Reward Modeling

Abstract

Reward models learned from human preferences are central to aligning large language models (LLMs) via reinforcement learning from human feedback, yet they are often vulnerable to reward hacking due to noisy annotations and systematic biases such as response length or style. We propose Bayesian Non-Negative Reward Model (BNRM), a principled reward modeling framework that integrates non-negative fac

Related papers

Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).