Starting from the original MDLM (Masked Discrete Diffusion Language Model) LLaDA-8B-Instruct, we trained the EvoTokenDLM LoRA adapter using the Continuous Trajectory Supervision method.
Our implementation replaces traditional hard binary masks with evolving soft token distributions. This allows EvoTokenDLM to facilitate a progressive transition from masked states to discrete outputs, effectively supporting revisable decoding.
For detailed instructions and code, please refer to the official GitHub repository:
EvoTokenDLM GitHub Repository1@article{zhong2026beyond,
2 title={Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models},
3 author={Zhong, Linhao and Wu, Linyu and Fang, Bozhen and Feng, Tianjian and Jing, Chenchen and Wang, Wen and Zhang, Jiaheng and Chen, Hao and Shen, Chunhua},
4 journal={arXiv preprint arXiv:2601.07351},
5 year={2026}
6}
