Compressor-VLA: Instruction-Guided Visual Token Compression for Efficient Robotic Manipulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Juntao, Ye, Feiyang, Zhang, Jing, Qian, Wenjing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911283575521280
author Gao, Juntao
Ye, Feiyang
Zhang, Jing
Qian, Wenjing
author_facet Gao, Juntao
Ye, Feiyang
Zhang, Jing
Qian, Wenjing
contents Vision-Language-Action (VLA) models have emerged as a powerful paradigm in Embodied AI. However, the significant computational overhead of processing redundant visual tokens remains a critical bottleneck for real-time robotic deployment. While standard token pruning techniques can alleviate this, these task-agnostic methods struggle to preserve task-critical visual information. To address this challenge, simultaneously preserving both the holistic context and fine-grained details for precise action, we propose Compressor-VLA, a novel hybrid instruction-conditioned token compression framework designed for efficient, task-oriented compression of visual information in VLA models. The proposed Compressor-VLA framework consists of two token compression modules: a Semantic Task Compressor (STC) that distills holistic, task-relevant context, and a Spatial Refinement Compressor (SRC) that preserves fine-grained spatial details. This compression is dynamically modulated by the natural language instruction, allowing for the adaptive condensation of task-relevant visual information. Experimentally, extensive evaluations demonstrate that Compressor-VLA achieves a competitive success rate on the LIBERO benchmark while reducing FLOPs by 59% and the visual token count by over 3x compared to its baseline. The real-robot deployments on a dual-arm robot platform validate the model's sim-to-real transferability and practical applicability. Moreover, qualitative analyses reveal that our instruction guidance dynamically steers the model's perceptual focus toward task-relevant objects, thereby validating the effectiveness of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18950
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Compressor-VLA: Instruction-Guided Visual Token Compression for Efficient Robotic Manipulation
Gao, Juntao
Ye, Feiyang
Zhang, Jing
Qian, Wenjing
Robotics
Computer Vision and Pattern Recognition
Machine Learning
Vision-Language-Action (VLA) models have emerged as a powerful paradigm in Embodied AI. However, the significant computational overhead of processing redundant visual tokens remains a critical bottleneck for real-time robotic deployment. While standard token pruning techniques can alleviate this, these task-agnostic methods struggle to preserve task-critical visual information. To address this challenge, simultaneously preserving both the holistic context and fine-grained details for precise action, we propose Compressor-VLA, a novel hybrid instruction-conditioned token compression framework designed for efficient, task-oriented compression of visual information in VLA models. The proposed Compressor-VLA framework consists of two token compression modules: a Semantic Task Compressor (STC) that distills holistic, task-relevant context, and a Spatial Refinement Compressor (SRC) that preserves fine-grained spatial details. This compression is dynamically modulated by the natural language instruction, allowing for the adaptive condensation of task-relevant visual information. Experimentally, extensive evaluations demonstrate that Compressor-VLA achieves a competitive success rate on the LIBERO benchmark while reducing FLOPs by 59% and the visual token count by over 3x compared to its baseline. The real-robot deployments on a dual-arm robot platform validate the model's sim-to-real transferability and practical applicability. Moreover, qualitative analyses reveal that our instruction guidance dynamically steers the model's perceptual focus toward task-relevant objects, thereby validating the effectiveness of our approach.
title Compressor-VLA: Instruction-Guided Visual Token Compression for Efficient Robotic Manipulation
topic Robotics
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2511.18950