RL for LLM reasoning acts as sparse policy selection at high-entropy tokens already present in the base model, enabling ReasonMaxxer—an efficient contrastive method that recovers most RL gains at three orders of magnitude lower cost.
What is the objective of reasoning with reinforcement learning?arXiv preprint arXiv:2510.13651
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
POCA combines Pareto optimization with curriculum alignment to improve multi-reward reinforcement learning for visual text generation without relying on weighted sums.
A survey synthesizing representative advances, common themes, and open problems in high-dimensional statistics while pointing to key entry-point works.
citing papers explorer
-
Rethinking RL for LLM Reasoning: It's Sparse Policy Selection, Not Capability Learning
RL for LLM reasoning acts as sparse policy selection at high-entropy tokens already present in the base model, enabling ReasonMaxxer—an efficient contrastive method that recovers most RL gains at three orders of magnitude lower cost.
-
POCA: Pareto-Optimal Curriculum Alignment for Visual Text Generation
POCA combines Pareto optimization with curriculum alignment to improve multi-reward reinforcement learning for visual text generation without relying on weighted sums.
-
High-Dimensional Statistics: Reflections on Progress and Open Problems
A survey synthesizing representative advances, common themes, and open problems in high-dimensional statistics while pointing to key entry-point works.