[2606.02684] Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation

Read full story on arxiv.org
Share
[2606.02684] Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation
AI disclosure

Summary

Abstract page for arXiv paper 2606.02684: Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation

Original reporting

Open original source
Read full article on arxiv.org