Memory Pretraining for Vision-Language-Action Model

Preprint Aug 2026

StellaVLA: In-Context Structured Demonstration for Generalizable Vision-Language-Action Models

The real-robot benchmark demonstrates that StellaVLA can use both human/robot demos and human-to-robot (XR) demos as in-context structured demonstration to help VLA model adapt to OOD tasks.

Siyu Xu, Yunke Wang, Zijian Wang et al. · 0 citations

Preprint Aug 2026

Decoding Task Progress from VLA Representations

The results suggest that VLAs have rich, linearly readable internal representations of semantic quantities like task progress, and that learning to read these signals offers a lightweight, interpretable path toward monitoring deployed visuomotor policies.

Atiksh Bhardwaj, E. W. Duan, Prithwish Dan et al. · 0 citations

Preprint Aug 2026

How Should Vision-Language-Action Models Use Proprioceptive State?

Five representative interfaces are implemented -- discrete state prompt, VLM prefix, action prefix, state expert, and feature modulation -- under matched implementation details, and evaluated on 45 atomic tasks spanning three task families plus 20 composite tasks.

Yiren Zhao, Ziyang Chen, Ziyang Rao et al. · 0 citations

Jul 2026

TFP: Temporally Conditioned Memory-Fusion Policies for Visuomotor Learning

Temporally Conditioned Memory-Fusion Policies (TFP), a lightweight memory-action framework for VLA backbones, is introduced and suggests that compact, event-sensitive memory dynamics can improve VLA policies under occlusion, visual perturbation, and stage-dependent task structure.

Yushen Liang, Yue Peng, Baosheng Jin et al. · arXiv.org · 0 citations

Jul 2026

Semantic Anchoring for Robotic Action Representations

This work examines whether a robot's action representations preserve the semantic structure captured by pretrained encoders and introduces a plug-and-play method that anchors action representations to a semantic manifold while decomposing representations into a shared semantic channel and a private channel, all discarded at inference, leaving the deployed model unchanged.

Yuan Xu, Youheng Shi, Chengyang Li et al. · arXiv.org · 0 citations

Preprint Jul 2026

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation

LaMem-VLA is introduced, a latent-memory-native framework that reconstructs historical experience into latent memory tokens and directly interweaves them with VLA reasoning, and enables memory to directly participate in VLA reasoning and guide action generation under a bounded context.

Hongyu Qu, Jianzhe Gao, Xiaobin Hu et al. · 1 citation

Memory Pretraining for Vision-Language-Action Model

Similar papers

StellaVLA: In-Context Structured Demonstration for Generalizable Vision-Language-Action Models

Decoding Task Progress from VLA Representations

How Should Vision-Language-Action Models Use Proprioceptive State?

TFP: Temporally Conditioned Memory-Fusion Policies for Visuomotor Learning

Semantic Anchoring for Robotic Action Representations

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation