ArgMaxRL: unbiased gradient estimator jointly optimizes all best@k objectives to fix RL diversity collapse

fpedregosa · x · 2026-09-02

Standard post-training RL objectives like GRPO optimize the expected quality of a single sample, causing mode collapse: greedy accuracy rises but the model can't produce diverse high-quality outputs, wasting test-time scaling potential.

ArgMaxRL, a new gradient estimator building on MaxRL and generalizing it from binary to continuous rewards:

The motivation: agentic systems tackling expert-level problems care about best@k, not top-1, and this work bakes test-time scaling directly into the training objective.

Original post →

More from Models

Models channel →