Author

Yi Ding

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.

#machine learning Preprint Aug 2026

Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement

It is found that learning concentrates on low log-probability tokens, and using a single fixed negative advantage matches the performance of teacher-provided ones, suggesting that OPD works largely by suppressing low log-probability tokens, which requires no teacher.

Yi Ding, Ruqi Zhang · 0 citations