@article{dtreerpotowardsmorereliablepolicyoptimiz, title = {d-TreeRPO: Towards More Reliable Policy Optimization for Diffusion Language Models}, author = {Leyi Pan and Shuchang Tao and Yunpeng Zhai and Zheyu Fu and Liancheng Fang and Minghua He and Lingzhe Zhang and Zhaoyang Liu and Bolin Ding and Aiwei Liu and Lijie Wen}, year = {2025}, eprint = {2512.09675}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2512.09675}, }