Environment-Aware Adaptive Pruning with Interleaved Inference Orchestration for Vision-Language-Action Models

TL;DR

EcoVLA is a training-free, adaptive pruning framework for Vision-Language-Action models that dynamically adjusts sparsity based on environment changes, significantly improving inference speed with minimal accuracy loss.

Contribution

EcoVLA introduces a novel environment-aware adaptive pruning method combined with interleaved inference orchestration, enabling real-time VLA model acceleration without retraining.

Findings

Achieves up to 2.18× speedup with only 0.5% success rate drop.

Effectively adapts to environment changes during VLA inference.

Validated on diverse models, benchmarks, and real-world robots.

Abstract

While Vision-Language-Action (VLA) models hold promise in embodied intelligence, their large parameter counts lead to substantial inference latency that hinders real-time manipulation, motivating parameter sparsification. However, as the environment evolves during VLA execution, the optimal sparsity patterns change accordingly. Static pruning lacks the adaptability required for environment dynamics, whereas fixed-interval dynamic layer pruning suffers from coarse granularity and high retraining overheads. To bridge this gap, we propose EcoVLA, a training-free, plug-and-play adaptive pruning framework that supports orthogonal combination with existing VLA acceleration methods. EcoVLA comprises two components: Environment-aware Adaptive Pruning (EAP) and Interleaved Inference Orchestration ($I^2O$). EAP is a lightweight adaptive channel pruning method that incorporates the temporal consistency of the physical environment to update sparsity patterns. $I^2O$ leverages the FLOPs bubbles inherent in VLA inference to schedule the pruning method in parallel, ensuring negligible impact on latency. Evaluated on diverse VLA models and benchmarks, EcoVLA delivers state-of-the-art performance, achieving up to 1.60$\times$ speedup with only a 0.4% drop in success rate, and further reaches 2.18$\times$ speedup with only a 0.5% degradation when combined with token pruning. We further validate the effectiveness of EcoVLA on real-world robots.