Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ye, Yifan, Ma, Jiaqi, Cen, Jun, Lu, Zhihe
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911311601860608
author Ye, Yifan
Ma, Jiaqi
Cen, Jun
Lu, Zhihe
author_facet Ye, Yifan
Ma, Jiaqi
Cen, Jun
Lu, Zhihe
contents Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant challenges for real-time deployment, as inference becomes computationally expensive and latency-sensitive in dynamic environments. To address this, we propose Token Expand-and-Merge-VLA (TEAM-VLA), a training-free token compression framework that accelerates VLA inference while preserving task performance. TEAM-VLA introduces a dynamic token expansion mechanism that identifies and samples additional informative tokens in the spatial vicinity of attention-highlighted regions, enhancing contextual completeness. These expanded tokens are then selectively merged in deeper layers under action-aware guidance, effectively reducing redundancy while maintaining semantic coherence. By coupling expansion and merging within a single feed-forward pass, TEAM-VLA achieves a balanced trade-off between efficiency and effectiveness, without any retraining or parameter updates. Extensive experiments on LIBERO benchmark demonstrate that TEAM-VLA consistently improves inference speed while maintaining or even surpassing the task success rate of full VLA models. The code is public available on \href{https://github.com/Jasper-aaa/TEAM-VLA}{https://github.com/Jasper-aaa/TEAM-VLA}
format Preprint
id arxiv_https___arxiv_org_abs_2512_09927
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models
Ye, Yifan
Ma, Jiaqi
Cen, Jun
Lu, Zhihe
Robotics
Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant challenges for real-time deployment, as inference becomes computationally expensive and latency-sensitive in dynamic environments. To address this, we propose Token Expand-and-Merge-VLA (TEAM-VLA), a training-free token compression framework that accelerates VLA inference while preserving task performance. TEAM-VLA introduces a dynamic token expansion mechanism that identifies and samples additional informative tokens in the spatial vicinity of attention-highlighted regions, enhancing contextual completeness. These expanded tokens are then selectively merged in deeper layers under action-aware guidance, effectively reducing redundancy while maintaining semantic coherence. By coupling expansion and merging within a single feed-forward pass, TEAM-VLA achieves a balanced trade-off between efficiency and effectiveness, without any retraining or parameter updates. Extensive experiments on LIBERO benchmark demonstrate that TEAM-VLA consistently improves inference speed while maintaining or even surpassing the task success rate of full VLA models. The code is public available on \href{https://github.com/Jasper-aaa/TEAM-VLA}{https://github.com/Jasper-aaa/TEAM-VLA}
title Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models
topic Robotics
url https://arxiv.org/abs/2512.09927