OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment

作者：T. Liu, Ran Xu, T.C.B. Yu, Ilgee Hong, Carl Yang, Tuo Zhao, Haoyu Wang · 年份：2026 · DOI：10.18653/v1/2026.acl-long.791 · 被引用次数：1 · 研究领域：Natural Language Processing Techniques

Reward modeling lies at the core of reinforcement learning from human feedback (RLHF), yet most existing reward models rely on scalar or pairwise judgments that fail to capture the multifaceted nature of human preferences.Recent studies have explored rubrics-as-rewards (RaR) that uses structured criteria to capture multiple dimensions of response quality.However, producing rubrics that are both reliable and scalable remains a key challenge.In this work, we introduce OpenRubrics, a diverse, large-scale collection of (prompt, rubric) pairs for training rubric-generation and rubric-based reward models.To elicit discriminative and comprehensive evaluation signals, we introduce Contrastive Rubric Generation (CRG), which derives both hard rules (explicit constraints) and principles (implicit qualities) by contrasting preferred and rejected responses.We further remove noisy rubrics via preserving preference-label consistency.Across multiple reward-modeling benchmarks, our rubricbased reward model, RUBRIC-RM, surpasses strong size-matched baselines by 8.4%.These gains transfer to policy models on instructionfollowing and biomedical benchmarks.The model weights and datasets are publicly available at https://huggingface.co/OpenRubrics.* These authors contributed equally to this work, order was determined randomly (by rolling a die).