Encoded but Not in Control: Revealing the Grounding Gap in Vision-Language Robot Policies

arXiv:2610.06235 · cs.RO, cs.LG · Submitted 2026-10-05 · Read on arXiv

cs.RO, cs.LG

Submitted: 2026-10-05

Updated: 2026-10-05

Comments: 30 pages, 7 figures

Code: https://github.com/anonymous-report-421/eval-of-gpt-6-astra-as-policy

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

Terminology

• Grounding — 1× in this paper · explained in 3 episodes

Sources

• Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation
• GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
• $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
• RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
• Modality-Composable Diffusion Policy via Inference-Time Distribution-level Composition
• RynnVLA-002: A Unified Vision-Language-Action and World Model
• WorldVLA: Towards Autoregressive Action World Model
• GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulation
• GaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks
• RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation
• In-Context Robot Learning with VLM Agents
• When Vision Overrides Language: Evaluating and Mitigating Counterfactual Failures in VLAs
• LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
• Vidar: Embodied Video Diffusion Model for Generalist Manipulation
• Reflective Planning: Vision-Language Models for Multi-Stage Long-Horizon Robotic Manipulation
• Addressing the Orchestration Gap in Generalist Robots via Physical Agency
• DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
• Act3D: 3D Feature Field Transformers for Multi-Task Robotic Manipulation
• Not All Features Are Created Equal: A Mechanistic Study of Vision-Language-Action Models
• Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Models

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• ARC: A Reasoning Recipe for Robot Foundation Models
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies