@article{teachingrlagentstoactbettervlmasactionad,
  title = {Teaching RL Agents to Act Better: VLM as Action Advisor for Online Reinforcement Learning},
  author = {Xiefeng Wu and Jing Zhao and Shu Zhang and Mingyu Hu},
  year = {2025},
  eprint = {2509.21126},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2509.21126},
}