When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA
cs.CV, cs.RO
Submitted: 2026-10-04
Updated: 2026-10-04
Terminology
Sources
- RT-1: Robotics Transformer for Real-World Control at Scale
- OpenVLA: An Open-Source Vision-Language-Action Model
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching
- EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models
- Bridging the Semantic-Action Gap in Visual Token Pruning for Efficient VLA Inference
- Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
- Token Merging: Your ViT But Faster
- Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs
- Evaluating Real-World Robot Manipulation Policies in Simulation
- SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration
- Action-aware Dynamic Pruning for Efficient Vision-Language-Action Manipulation
- FoPru: Focal Pruning for Efficient Large Vision-Language Models
- Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retention
- OTPrune: Distribution-Aligned Visual Token Pruning via Optimal Transport
- Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimization
- SpecPrune-VLA: Accelerating Vision-Language-Action Models via Action-Aware Self-Speculative Pruning
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models