RLAVR uses the Corrective Advantage Gap metric and CARE policy to actively acquire ground-truth labels for key samples, stabilizing RLVR training and boosting performance with limited annotation budgets.
Prism: A unified framework for post-training llms without verifiable rewards.arXiv preprint arXiv:2601.04700, 2026
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
When Self-Belief Misleads: Active Label Acquisition for Reinforcement Learning with Verifiable Rewards
RLAVR uses the Corrective Advantage Gap metric and CARE policy to actively acquire ground-truth labels for key samples, stabilizing RLVR training and boosting performance with limited annotation budgets.