LayerRoute: Action-Conditioned Mixture-of-Layers Routing for Vision-Language-Action Policies

arXiv:2609.06079 · cs.AI, cs.LG · Submitted 2026-09-05 · Read on arXiv

cs.AI, cs.LG

Submitted: 2026-09-05

Updated: 2026-10-10

Comments: 15 pages, 7 figures, 16 tables, including appendix

Code: https://github.com/PhysicalIntelligence/openpi

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Ablation — 1× in this paper · explained in 1 episode
• Ablation Studies — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Vision-Language Models (VLMs) — 1× in this paper · explained in 6 episodes

Sources

• Qwen3-VL Technical Report
• Eliciting Latent Predictions from Transformers with the Tuned Lens
• PaliGemma: A versatile 3B VLM for transfer
• $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
• RT-1: Robotics Transformer for Real-World Control at Scale
• StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing
• Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets
• MiMo-Embodied: X-Embodied Foundation Model Technical Report
• Deep Networks with Stochastic Depth
• OTTER: A Vision-Language-Action Model with Text-Aware Visual Feature Extraction
• $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
• OpenVLA: An Open-Source Vision-Language-Action Model
• HAMLET: Switch your Vision-Language-Action Model into a History-Aware Policy
• Evaluating Real-World Robot Manipulation Policies in Simulation
• Planning Takes More Than Token Prediction: Causal Plan for Benchmarking and Building Physically Grounded Embodied Reasoners
• RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots
• SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model
• Mixture-of-Depths: Dynamically allocating compute in transformer-based language models
• MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulation
• Attention Residuals

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems