ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models

• URL: http://arxiv.org/abs/2605.29438v1
• Date: Thu, 28 May 2026 06:33:05 GMT
• Title: ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models
• Abstract summary: Vision-Language-Action (VLA) models are a powerful paradigm for generalist robotic control.<n> ElegantVLA is a plug-in phase-adaptive inference framework that accelerates VLA models through intra-model dynamic compute scheduling.<n>Experiments on GR00T and CogACT achieve up to 2.55x and 3.77x speedup, and on six real-world GR00T tasks ElegantVLA cuts computation by 2.18x while raising control frequency from 13.8 Hz to 26.3 Hz.
• Score: 46.57405778313275
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language-Action (VLA) models are a powerful paradigm for generalist robotic control. However, their high computational cost and limited control frequency hinder real-time robotic manipulation, especially when large vision-language backbones and iterative action heads run at every control step. Existing VLA acceleration methods often optimize individual components or rely on fixed acceleration rules, treating different control steps with largely fixed computation and overlooking the non-uniform reasoning demands of sequential embodied control. Inspired by human motor control, where cognitive and feedback resources concentrate on goal-sensitive stages, we argue that VLA models should learn when to invest full computation and when to reuse prior computation. We propose ElegantVLA, a plug-in phase-adaptive inference framework that accelerates VLA models through intra-model dynamic compute scheduling. ElegantVLA introduces a lightweight scheduler that observes temporal representation similarity, robot-motion cues, and episode progress to jointly allocate computation across the vision encoder, LLM, and action head. For perception-language reasoning, the scheduler selects a five-level Vision-LLM compute mode, from full recomputation to multi-step temporal reuse, based on visual-language representation stability. For action generation, it selects a three-level denoising mode, reusing intermediate denoising states during stable motion while preserving full refinement for goal-sensitive stages. By coordinating these decisions, ElegantVLA offers a general acceleration framework for modern VLA pipelines with explicit action-generation modules, without modifying or retraining the base model. Experiments on GR00T and CogACT achieve up to 2.55x and 3.77x speedup, and on six real-world GR00T tasks ElegantVLA cuts computation by 2.18x while raising control frequency from 13.8 Hz to 26.3 Hz.

Related papers

• AdaVLA: Adaptive Step Flow Matching for Training-free Acceleration of Vision-Language-Action Models [1.0323063834827415]
  We propose AdaVLA, an online, training-free adaptive framework for fast yet accurate flow-matching-based Vision-Language-Action models.<n>We show that our method achieves $1.87times$ and $2.24times$ speedups for $_0.5$ and X-VLA, respectively, with negligible degradation in success rates.
  arXiv  Detail & Related papers  (2026-08-29T11:44:18Z)
• Global Prior Meets Local Consistency: Dual-Memory Augmented Vision-Language-Action Model for Efficient Robotic Manipulation [95.89924101984566]
  We introduce OptimusVLA, a dual-memory VLA framework with Global Prior Memory (GPM) and Local Consistency Memory (LCM)<n>GPM replaces Gaussian noise with task-level priors retrieved from semantically similar trajectories.<n>LCM injects a learned consistency constraint that enforces temporal coherence and smoothness of trajectory.
  arXiv  Detail & Related papers  (2026-02-22T15:39:34Z)
• Learning to Accelerate Vision-Language-Action Models through Adaptive Visual Token Caching [25.295588774254952]
  We reformulate inference acceleration as a learnable policy optimization problem.<n>We propose a novel framework that integrates a task-aware decision-making process directly into the Vision-Language-Action model.<n>Our method achieves a 1.76x wall-clock inference speedup while simultaneously improving the average success rate.
  arXiv  Detail & Related papers  (2026-01-31T12:12:51Z)
• DynamicVLA: A Vision-Language-Action Model for Dynamic Object Manipulation [52.83157499300261]
  We present DynamicVLA, a framework for dynamic object manipulation that integrates temporal reasoning and closed-loop adaptation.<n>We introduce the Dynamic Object Manipulation benchmark, built from scratch with an auto data collection pipeline.<n>Extensive evaluations demonstrate remarkable improvements in response speed, perception, and generalization.
  arXiv  Detail & Related papers  (2026-01-29T18:59:51Z)
• AC^2-VLA: Action-Context-Aware Adaptive Computation in Vision-Language-Action Models for Efficient Robotic Manipulation [21.23747444669735]
  We propose Action-Context-aware Adaptive Computation for VLA models (AC2-VLA)<n>AC2-VLA adaptively performs cognition reuse across timesteps, token pruning, and selective execution of model components within a unified mechanism.<n>Experiments on robotic manipulation benchmarks show that AC2-VLA achieves up to a 1.79times speedup while reducing FLOPs to 29.4% of the dense baseline.
  arXiv  Detail & Related papers  (2026-01-27T14:10:39Z)
• ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation [52.94334113271359]
  ActiveVLA is a vision-language-action framework that empowers robots with active perception capabilities for high-precision, fine-grained manipulation.<n>We show that ActiveVLA achieves precise 3D manipulation and outperforms state-of-the-art baselines on three simulation benchmarks.
  arXiv  Detail & Related papers  (2026-01-13T08:29:07Z)
• Asynchronous Fast-Slow Vision-Language-Action Policies for Whole-Body Robotic Manipulation [10.09057399213028]
  Vision-Language-Action (VLA) systems integrate a Vision-Language Model (VLM) for semantic reasoning with an action expert generating continuous action signals.<n>We introduce a truly asynchronous Fast-Slow VLA framework (DuoCore-FS) to organize the system into a fast pathway for action generation and a slow pathway for rich VLM reasoning.
  arXiv  Detail & Related papers  (2025-12-23T09:28:20Z)
• dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought [66.78110237549087]
  Vision-Language-Action (VLA) models are emerging as a next-generation paradigm for robotics.<n>We introduce dVLA, a diffusion-based VLA that unifies visual perception, language reasoning, and robotic control in a single system.
  arXiv  Detail & Related papers  (2025-09-30T02:36:11Z)
• SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration [70.72227437717467]
  Vision-Language-Action (VLA) models have attracted increasing attention for their strong control capabilities.<n>Their high computational cost and low execution frequency hinder their suitability for real-time tasks such as robotic manipulation and autonomous navigation.<n>We propose SP-VLA, a unified framework that accelerates VLA models by jointly scheduling models and pruning tokens.
  arXiv  Detail & Related papers  (2025-06-15T05:04:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.