Paper page - REA: A Routed Embodied Agent Runtime for Asynchronous Brain-Cerebellum Control

Source: https://huggingface.co/papers/2610.42053

Abstract

REA unifies deliberative and reactive policies in a single runtime with asynchronous coordination, achieving high success rates with reduced control overhead across simulated and real robot tasks.

Foundation modelsare becoming remarkably reliable at high-level reasoning, yet they remain too slow and costly to serve as fine-grained controllers. Conversely, fast policies likeVLAsandwhole-body controllersoffer more efficient reflexes, but struggle to generalize. While pairing a deliberative brain (foundation models) with a reactive cerebellum (fast policies) seems natural for embodied systems, two obstacles remain: methods are built and scored on incompatible stacks, and their asynchronous coordination is difficult. For example, an agent that calls the VLA in the agentic loop can stall the fast execution, while a VLA that calls the agent on errors tends to re-trigger fallbacks repeatedly or completely miss silent failures. To address this, we propose REA (Routed Embodied Agent), a unified runtime where a deliberativeSystem 2, an embodied routerSystem 1and reactiveSystem 0policies cooperate under a single golden rule - no running policy waits on the brain. Instead of querying the brain at each failure, REA employs aroute memorythat stores each policy’s measured success ahead of time, a boundary-triggeredjudge gatethat consults the brain on policy execution, and a shared-budgetescalation ladderfor bounded recovery. Evaluated on 3,000 test episodes across 60LIBERO-Protasks, REA reaches 95.9% success and beats the best single-agent baseline in its own policy pool (94.7%) while reducing control steps by 25% and control time by 17%. In a separate 100-episodeRoboTwin 2.0study, REA achieves 84% success with VLA policies and 86% with VLA+code, compared with 80% for the strongest single VLA baseline MEDex. These results demonstrate learned and programmatic policies operating within the same runtime. REA also transfers toBEHAVIOR-1Kand a real Unitree G1 humanoid, achieving the highest observed success among the compared methods.

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2610\.42053

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.42053 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.42053 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.42053 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.