Efficient VLAs
VLA: Vision-Language-Action Model
Overview
Efficient VLA research studies how robots and other embodied agents can connect visual observations and language instructions to actions under limited compute, memory, and latency budgets. Common approaches include compact backbones and action heads, visual-token pruning, quantization and cache reuse, and faster action generation through chunking, parallel decoding, or fewer diffusion steps. Other methods adapt computation to the task or reuse compressed memory so models can respond quickly while retaining context for long-horizon manipulation, navigation, and driving. Results on benchmarks such as LIBERO, RoboCasa, and driving and real-robot evaluations show that many methods can improve speed or reduce resource use, but efficiency must also be judged by task success, robustness, motion quality, and end-to-end execution cost.
Latest papers
110 papers9 papers in the last four weeks, down 50% on the four weeks before. 0.1% of all new papers.
XS-VLA: Teaching Tiny Vision-Language-Action Models with Spatial Supervision and Demonstration Conditioning
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
Z-1: Efficient Reinforcement Learning for Vision-Language-Action Models
Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?
UniFS: Unified Fast-to-Slow Hierarchical Architecture for Vision-Language-Action Models
PolicyTrim: Boosting Intrinsic Policy Efficiency of Vision-Language-Action Models
UniviewVLA: A Unified Multiview Vision-Language-Action Model with World Modeling
FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation
Finetuning Vision-Language-Action Models Requires Fewer Layers Than You Think
PearlVLA: Progressive Embodied Action-Plan Refinement in Latent Space
Acting While Understanding: Asynchronous Semantic-Action Decoupling for Real-Time Vision-Language-Action Models
Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models
BLUE: Toward Better Language Use in Efficient Vision-Language-Action Models for Autonomous Driving
vla.cpp: A Unified Inference Runtime for Vision-Language-Action Models
TBD-VLA: Temporal Block Diffusion Vision Language Action Model
VisualThink-VLA: Visual Intermediate Reasoning for Effective and Low-Latency Vision-Language-Action Policies
ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models
Ω-QVLA: Robust Quantization for Vision-Language-Action Models via Composite Rotation and Per-step Scaling
EXPO-FT: Sample-Efficient Reinforcement Learning Finetuning for Vision-Language-Action Models
DriveMA: Rethinking Language Interfaces in Driving VLAs with One-Step Meta-Actions
ActQuant: Sub-4-bit Action-Guided Backbone Weight-Only Quantization for Vision-Language-Action Models
Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation
libero\_goal. Additional analysis indicates that phase-level supervision and multi-frame directional cues make the student less sensitive to noisy teacher actions, such as erroneous high-frequency gripper changes. Overall, VLA-AD demonstrates that offline semantic guidance from VLMs can substantially improve the efficiency, robustness, and deployability of VLA policy distillation.