@article{ttvsboostingselfexploringreinforcementle,
  title = {TTVS: Boosting Self-Exploring Reinforcement Learning via Test-time Variational Synthesis},
  author = {Sikai Bai and Haoxi Li and Jie Zhang and Yongjiang Liu and Song Guo},
  year = {2026},
  eprint = {2604.08468},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.08468},
}