AUG 10, 2026 · PREPRINT
Beyond Solvability: Task Learnability as a Static Prior for LLM RL Post-Training
arXiv
Early-stage empirical work proposing a new method for task selection in LLM post-training, demonstrated on benchmarks but without peer review, hard endpoints, or evidence of deployment impact.
Study details
DesignComparative empirical study with benchmark…
InterventionTrajVal task-sampling prior based on estima…
ComparatorUniform task sampling and existing online s…