@article{blockwiseadvantageestimationformultiobje,
  title = {Blockwise Advantage Estimation for Multi-Objective RL with Verifiable Rewards},
  author = {Kirill Pavlenko and Alexander Golubev and Simon Karasik and Boris Yangel},
  year = {2026},
  eprint = {2602.10231},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.10231},
}