What is the impact of VLA model scale (7B vs 13B) on object grounding accuracy and path completion rate in Lon

Embodied AI is widely recognized as a cornerstone of artificial general intelligence (AGI) because it involves controlling embodied agents to perform tasks in the physical world. Building on the success of large language models (LLMs) and vision-language models (VLMs), a new category of multimodal models-referred to as vision-language-action (VLA) models-has emerged to address language-conditioned robotic tasks in embodied AI by leveraging their distinct ability to generate actions. The recent proliferation of VLAs necessitates a comprehensive survey to capture the rapidly evolving landscape.

Research goal: What is the impact of VLA model scale (7B vs 13B) on object grounding accuracy and path completion rate in LongNav-R1 on R2R-CE under varying instruction complexity?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 9.0/10.