SOURCE-LINKED INTELLIGENCE

Why Sample What You Can Enumerate? Exact Policy Optimization for Genomic Tool Selection

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Reinforcement learning over a frozen reasoner has become a common recipe for teaching a policy which external tools to invoke. We show that this recipe becomes structurally mismatched in specialist scientific settings where the complete tool-subset space is enumerable. There, a small set of recurring computational capabilities covers the domain, so the space of tool subsets is combinatorial yet small enough to enumerate, and GRPO still estimates an action expectation from a handful of sampled rollouts. Worse, the approximation degrades as training succeeds: as the policy concentrates on prefer

Read original source ↗ Open in workspace

recordType

paper

region

Global