On-policy distillation (OPD) is increasingly applied to frontier foundation model post-training. Prior work in this area has largely focused on algorithmic advances, yet it remains unclear how much OPD depends on its training questions and, in particular, how far this dependence can be reduced. Across two representativ...
Gengsheng Li, Mao Zheng, Ming-Yang Song et al.· 0 citations
TrustMOPD replaces example-level teacher selection with label-free, token-level supervision allocation and closes the overall-score gap between the initial student and oracle-routed teachers when trained on single- and multi-teacher on-policy distillation.
Jie Sun, Mao Zheng, Ming-Yang Song et al.· 5 citations
DF-OPD matches and even surpasses real data, and the questions it produces track the teacher's own post-training data on three key diagnostics of training dynamics, which other real datasets do not.
Gengsheng Li, Mao Zheng, Ming-Yang Song et al.· 1 citation
Experiments on reasoning, code-generation, scientific-knowledge, scientific-knowledge, and tool-use benchmarks show that these implementations can be executed through the same verl-based backend while retaining their method-specific objectives and task-dependent performance profiles.
Jie Sun, Mao Zheng, Mingyang Song et al.· arXiv.org· 0 citations
Experiments show Influence-Directed Adaptive On-Policy Distillation (IDA-OPD), rather than relying on costly full-vocabulary Forward-KL objectives, preserves entropy-expanding updates while replacing entropy-contracting ones with divergence-adaptive advantage shrinkage, using only the teacher's sampled-token log-probab...
Run Yang, Runpeng Dai, Jie Sun et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
