Author

Shurun Yuan

View on Semantic Scholar

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#natural language process... May 2026

MAIGO: Mitigating Lost-in-Conversation with History-Cleaned On-Policy Self-Distillation

Results show that self-contamination is a trainable component of the LiC gap, and propose MAIGO, an on-policy self-distillation method that reduces this contamination using history-cleaned references from the model's own policy.

Hao Zheng, Yun Zhu, Shurun Yuan et al. · arXiv.org · 3 citations