# RouterVLA: Budgeted Commissioning and Expert Onboarding for Growing VLA Pools

arXiv: https://arxiv.org/abs/2606.27355 · HTML: https://mlsanity.com/paper/2606.27355/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jun 2026.

## Built on by (most cited first, 2 of 2)

- [Think Only When Needed: Prompt-Authority Control for Selective Slow-Path Intervention in Vision-Language-Action Manipulation](https://mlsanity.com/paper/2608.23224/) (2026)
- [RoboAware: Learning to Coordinate Embodied Skills from Counterfactual Outcomes](https://mlsanity.com/paper/2610.11480/) (2026)

## Abstract (arXiv)

Robotic teams often maintain several vision-language-action policies but still deploy one global winner. We study two recurring decisions: which expert to deploy for a new condition and which candidate to add to the pool. RouterVLA combines a split-clean prior and outcome-disjoint probes with onboarding that credits only failures the incumbent pool cannot handle. Under an exactly cost-matched probe budget, it reaches 60.53% held-out success, a +1.64pp gain over a semantic shortlist. Both criteria independently converge on the same five experts, confirming that the candidates best positioned to cover the base pool's blind spots are also broadly capable.