@article{tdgrokkinglearningfromzerorewardproblems,
  title = {TD-Grokking: Learning from Zero-Reward Problems by Training-Time Decomposition},
  author = {Ningyuan Xi and Hao Xu and Hongsheng Xin and Ning Miao},
  year = {2026},
  eprint = {2606.09883},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.09883},
}