SOURCE-LINKED INTELLIGENCE
Reward-Adaptive Co-Evolution of Policy and Rubric
Rubric-based rewards provide fine-grained, interpretable supervision for open-ended reinforcement learning, where holistic scalar rewards struggle to capture multidimensional response quality. Static rubrics can lose discriminative power as the policy distribution evolves, while existing policy--rubric co-evolution methods rely on globally prescribed update schedules that ignore heterogeneous reward informativeness. We introduce RACE, a Reward-Adaptive Co-Evolution framework. At the group level, per-rollout-group training-signal routing classifies groups as informative, diversity-collapsed, or
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-10-29T14:00:07.000Z
First collected: 2026-10-31T14:12:28.156Z. This is not the publication date.