RAPID: Redundancy-Aware and Compatibility-Optimal Edge-Cloud Partitioned Inference for Diverse VLA Models

TL;DR

RAPID is a novel edge-cloud inference framework designed for Vision Language Action models, reducing inference costs and improving robustness by addressing noise interference and redundancy, achieving up to 1.73x speedup with minimal overhead.

Contribution

The paper introduces RAPID, a new ECC inference framework that optimally handles redundancy and noise in VLA models for embodied intelligence.

Findings

Achieves up to 1.73x speedup in inference

Maintains only 5-7% overhead

Effectively handles visual noise and redundancy

Abstract

Vision Language Action (VLA) models are mainstream in embodied intelligence but face high inference costs. Edge-Cloud Collaborative (ECC) inference offers an effective fix by easing edge-device computing pressure to meet real-time needs. However, existing ECC frameworks are suboptimal for VLA models due to two challenges: (1) Mainstream environment-oriented edge-cloud partitioning methods are susceptible to interference from visual noise; (2) Existing edge-cloud partitioning methods overlook the step-wise redundancy unique to embodied tasks, thereby disrupting the physical continuity of motion. To address these issues, we propose a novel ECC inference framework, termed RAPID. Specifically, we developed an implementation tailored to the proposed framework. Experiments demonstrate this achieves a speedup of up to 1.73x with only 5%~7% overhead.