Skip to content
Preprint

Generalists Act, Specialists Intervene: Modular Stage-Selective Reinforcement Learning for Vision-Language-Action Manipulation

Sep 2026 · 0 citations · 37 references
Computer Science

TL;DR

This paper introduces RouteRLT, a modular framework that coordinates a generalist VLA, used as the default controller, with designated precision-critical RL specialists, and trains a phase-aware coordination mechanism that handles handoffs between the generalist and the specialists.

Abstract

Vision-language-action (VLA) models often struggle in the precision-critical phases of multi-stage manipulation tasks. To mitigate this issue, VLA models can be used in conjunction with reinforcement learning (RL) specialists that are specifically trained to handle the precision-critical phases. However, the coordination between the base VLA model and the RL specialists, which dictates when a specialist should take over from the base VLA and vice-versa, remains an open research question. In this paper, we address this gap by introducing RouteRLT, a modular framework that coordinates a generalist VLA, used as the default controller, with designated precision-critical RL specialists. At a high level, our framework trains a phase-aware coordination mechanism that handles handoffs between the generalist and the specialists. We evaluate RouteRLT on the LIBERO and LIBERO-Plus benchmarks, as well as on a physical connector pickup and insertion task. Overall, we find that RouteRLT improves success on LIBERO, and retains net gains on LIBERO-Plus. On the physical task, RouteRLT completes 65.7% of trials, compared with 8.6% for the baseline. Altogether, these results demonstrate that learned coordination builds on generalist VLA capabilities to improve task completion in precision-critical manipulation.

View source

Similar papers

Sep 2026

Dissecting Advantage-Guided Post-Training for Vision-Language-Action Policies

This work dissects advantage-guided VLA post-training through a controlled empirical study that separates these design choices while accounting for their distinct estimands, and develops stage-specific offline evaluation methods to screen alternative choices efficiently, without requiring extensive real-robot policy ev...

Jia-Hang Cao, Han-Ye Zhao, Hang Lai et al. · 0 citations
Preprint Oct 2026

Triggering Generalist Reasoning via Predictive Uncertainty for Dual-System VLA

Dual-system Vision-Language-Action (VLA) models improve real-time robotic control by pairing a slow, reasoning-capable generalist with a fast specialist action expert. However, existing methods invoke the generalist at a fixed frequency, ignoring the fact that decision-making complexity varies throughout a rollout. Thi...

Hyemin Yang, Wooseong Jeong, Giwon Lee et al. · 1 citation · ⚡1
Preprint Oct 2026

Axiom-0: Hierarchical Self-Refinement for Vision-Language-Action Policies

Axiom-0, a VLA policy that addresses both the quality of its representations and their use in action generation through hierarchical self-refinement, is presented, a VLA model that combines pretrained multimodal understanding with learned control on unfamiliar task compositions.

Ren-Bin Shu, Jing Yang, Xiang-Yuan You et al. · 0 citations
Sep 2026

Find Something You Can't Do: Agentic Real-World Reinforcement Learning for Self-Improving VLA Models

FIND is introduced, an agentic real-world RL framework that closes the loop between scene understanding, weakness-aware practice, self-evaluation, and policy improvement in a persistent workspace and reframes autonomous practice as a scene-conditioned, performance-aware task-selection problem.

Yuan Fang, Ze-Chu Li, Hao-Lei Tong et al. · 4 citations · ⚡1
Preprint Oct 2026

PRISM: Precision Forcing for Generalist Vision-Language-Action Policies

Simulation and real-robot experiments show that PRISM preserves or improves general-task performance while raising mean end-to-end success from 38.3% to 86.7% across drive connector insertion, bimanual connector mating, and screw tightening, substantially extending the precision boundary of existing embodied policies w...

Cheng-Hao Yue, Pei-Yan Li, Hao Wang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.