@article{speculatingllmschinesetrainingdatapollut, title = {Speculating LLMs' Chinese Training Data Pollution from Their Tokens}, author = {Qingjie Zhang and Di Wang and Haoting Qian and Liu Yan and Tianwei Zhang and Ke Xu and Qi Li and Minlie Huang and Hewu Li and Han Qiu}, year = {2025}, eprint = {2508.17771}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2508.17771}, }