Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training

arXiv:2605.12483v5 Announce Type: replace-cross Abstract: In settings where labeled verifiable training data is the binding constraint, each checked example should be allocated to the model and reward density where it is most informative.

← Back to all developments · Signals are a development reported by more than one source. See the methodology.