On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation
Paper • 2611.05609 • Published • 24
None defined yet.
OneTrans-GSR: A Generative Slate Ranking Paradigm Unifying Reranking and Mix-Ranking with One Autoregressive Transformer
On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation