@article{grailgradientreweightedadvantagesforrein,
  title = {GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards},
  author = {Tej Deep Pala and Vernon Toh and Soujanya Poria},
  year = {2026},
  eprint = {2606.04889},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.04889},
}