On-policy distillation (OPD) trains agents on their own trajectories using supervision from a stronger teacher. We uncover a counterintuitive phenomenon in multi-turn agentic tasks: a smaller, initially weaker student can substantially outperform its teacher during training, only to regress as distillation continues. W...
Ji-Ke Zhong, Ming Li, Yu-Xiang Lai et al.· 0 citations
This work establishes concept support, beyond individual sample importance, as a useful principle for data-efficient and robust learning, as a useful principle for data-efficient and robust learning.
Zi-Heng Zhang, Neha Hulkund, Colin Lee et al.· 0 citations
The results suggest context learning hinges on not only content acquisition but also specification acquisition, and designs a deliberately simple intervention PSCI (private specification-contract induction) which extracts local specifications and enforces them through adversarial checking and repair.
Jike Zhong, Ming Li, Yuxiang Lai et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.