FM-VLA: Force-based Memory for Vision-Language-Action Models in Contact-Rich Manipulation

Ruicheng Li Qi-Xiu Li Ruichun Ma Yu Deng Lin Luo Zhi-Ying Du Jianfeng Xiang Hui-Zhi Liang Ruicheng Wang Jiao-Long Yang Bai-Ning Guo

Jul 2026 · arXiv.org · Vol abs/2607.18231 · 1 citation · 40 references

Computer Science

TL;DR

FM-VLA is proposed, a VLA model with force-based memory, enabling temporal context reasoning for non-Markovian, contact-rich manipulation, and achieves over 80% success rate with minimal inference overhead, significantly outperforming baseline approaches.

Abstract

Vision-language-action (VLA) models have achieved impressive generalization in robotic manipulation, and recent memory-augmented VLAs have relaxed the Markovian assumption by conditioning on past images or language summaries. Vision-based memory approaches address this by conditioning on sampled past image frames, but they are computationally expensive and fundamentally limited when temporal events are visually ambiguous, e.g., pushing a button multiple times with small movements. We propose FM-VLA, a VLA model with force-based memory, enabling temporal context reasoning for non-Markovian, contact-rich manipulation. We encode force histories into compact force memory tokens with a variational autoencoder (VAE) pretrained with force time series reconstruction. By projecting force latent representations and short state history as additional conditioning tokens to the action expert module, we enable VLAs to leverage accumulated contact event history to guide manipulation. We evaluate FM-VLA on three memory-dependent tasks, including finding a hidden block, pressing a button, and wiping a dish for a specific number of times. Our lightweight force memory achieves over 80% success rate with minimal inference overhead, significantly outperforming baseline approaches. Project page: https://qft-333.github.io/FM-VLA-Page/

DOI arXiv Semantic Scholar

View source

Similar papers

Jul 2026

FibVLA: An Efficient Temporal Vision-Language-Action Model with Fibonacci Sampling

FibVLA is introduced, an efficient framework featuring temporal perception of long-context history and logarithmic hindsight sampling to both proprioceptive states and visual frames to capture long-term temporal dependencies with minimal redundancy.

Li Lin, Wujun Xu, Wei Meng et al. · arXiv.org · 1 citation

Jul 2026

τ: Learning Touch-Augmented Vision-Language-Action Models from Future Visual Supervision

Tau is presented, a touch-augmented VLA framework that learns an action-conditioned spatiotemporal tactile representation from future visual supervision inspired by the Joint-Embedding Predictive Architecture (JEPA), and fuses it with vision-language features for action generation.

Ning Cheng, Jinan Xu, Wanlin Li et al. · arXiv.org · 1 citation

Jul 2026

MemVLN: Episodic and Procedural Memory for Vision-and-Language Navigation

MemVLN is proposed, a novel VLN framework that achieves state-of-the-art performance with real-time inference efficiency (14 FPS) and introduces Procedural Memory for fast action with a compact vocabulary of atomic mid-level actions to bypass auto-regressive decoding latency.

Yuqi Liu, Sheng-Ju Qian, Tianyuan Qu et al. · arXiv.org · 0 citations

#computer vision Preprint Aug 2026

Explicit Language Memory for Long-Horizon Planning in Vision-Language-Action Models

The proposed hierarchical long-horizon VLA architecture with an explicit language-memory module improves the success rate and robustness of VLA models on complex long-horizon tasks while providing an interpretable semantic account of the decision process.

Hou-Ze Xu, Jizhong Li, Zi-Yi Ye · 0 citations

#artificial intelligence Preprint Aug 2026

PHR-VLA: Planning Horizon Reasoning for Vision-Language-Action Models

PHR-VLA introduces a lightweight auxiliary future head that, during training, aligns the VLA's internal representations with latent dynamics extracted from future observations, demonstrating that privileged latent dynamics alignment provides an effective training signal for improving anticipatory reasoning in VLA polic...

Davood Soleymanzadeh, Kai-Di Zhang, Zhi-Yuan Zhang et al. · 2 citations

Preprint Sep 2026

IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

IMLE-VLA is introduced, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE), and promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely.

Kian Hosseinkhani, Qin-He Peng, George Shramko et al. · 1 citation