VLA-ULAP: Interleaving Cloud VLA Calls with Ultra-Lightweight Local Action Prediction at the Edge
cs.RO, cs.LG
Submitted: 2026-09-16
Updated: 2026-09-26
Comments: Preprint
Code: https://github.com/TheRobotStudio/SO-ARM100
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Action Chunk Scheduling for Batched Robot Policy Serving
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- Dynamic Execution Commitment of Vision-Language-Action Models
- LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
- Denoising Tells When to Replan: Denoising-Variance Adaptive Chunking for Flow-Based Robot Policies
- Latent Bridge: Feature Delta Prediction for Efficient Dual-System Vision-Language-Action Model Inference
- Bridging the Semantic-Action Gap in Visual Token Pruning for Efficient VLA Inference
- Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs
- ActionCache: Training-Free Acceleration for Vision-Language-Action Models with Action Caching and Refinement
- VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon
- Latency-Tolerant Cloud-Edge Collaborative Vision-Language-Action Models via Emergent Representational Specialization
- Offload or Overload: A Platform Measurement Study of Mobile Robotic Manipulation Workloads
- VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model
- Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models
- VLA Knows Its Limits: Adaptive Execution Horizons for Robot Policies
- EagleVLA: Towards Onboard Real-Time Robot Control via Foresight-Aligned Asynchronous Inference
- Rethink Before You Execute: Adaptive Execution for World Action Models
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving