SOURCE-LINKED INTELLIGENCE

RLVR$^{2}$: Reinforcement Learning with Verifiable Rubric-based Ranking

arXiv · AI, language, vision and robotics · article · Sep 20, 2026 · UTC

Reinforcement Learning with Verifiable Rewards (RLVR) is expanding from tasks with well-defined correctness signals, such as mathematics and code, toward multifaceted quality requirements specified by multi-dimensional rubrics. Since policy optimization consumes one scalar per rollout, rubric-based pipelines must map multiple criterion scores into a scalar reward. This aggregation is often treated as score scaling, but it implicitly determines how quality dimensions trade off during training. The prevailing practice, normalizing each criterion and taking a linear combination, assumes that card

Read original source ↗ Open in workspace

recordType

paper

region

Global