Balancing Understanding and Generation in Discrete Diffusion Models
Yue Liu, Yuzhong Zhao, Zheyong Xie, Qixiang Ye, Jianbin Jiao, Yao Hu, Shaosheng Cao, Yunfan Liu · 2026
Read Balancing Understanding and Generation in Discrete Diffusion Models by Yue Liu, Yuzhong Zhao, Zheyong Xie, Qixiang Ye, Jianbin Jiao, Yao Hu, Shaosheng Cao, Yunfan Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01362.txt.
Description
Yue Liu and co-authors build XDLM, a discrete diffusion plan balancing sense and making. Masked diffusion plans grasp sense and zero-shot transfer well, while uniform-noise plans make few-step pictures well, yet neither balances both gifts. XDLM bridges them with a steady noise kernel, unifying both plans in theory as edge cases, and easing a memory pinch via posterior algebra. Tests push the sense-versus-make frontier forward, beating each rival on its weak side and scaling to an 8B-parameter model.
Catalog record
| CREATOR | Yue Liu, Yuzhong Zhao, Zheyong Xie, Qixiang Ye, Jianbin Jiao, Yao Hu, Shaosheng Cao, Yunfan Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 118690 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01362