Vision-Language-Action Models

Also known as VLA

These models combine vision-language backbones with action generators that produce continuous motion or discrete action tokens, often in chunks and with proprioceptive input. They are commonly trained on robot demonstrations, with flow-matching and diffusion policies, reinforcement fine-tuning, and predictive world models used to improve control and planning. Research spans tabletop and bimanual manipulation, mobile and aerial navigation, autonomous driving, and humanoid control, evaluated on benchmarks such as LIBERO, CALVIN, RoboTwin, and driving simulators. A central challenge is making broad visual and language competence translate into reliable closed-loop behavior under new objects, viewpoints, embodiments, delays, and safety constraints; benchmark success can still conceal weak instruction grounding and sim-to-real failures.

Latest papers 1,154

249 papers in the last four weeks, up 144% on the four weeks before. 1.5% of all new papers.

All topics

Decoupling Semantics and Geometric Grounding: Spatial Visual Prompts for Language-Conditioned Imitation Learning
Jun 24, 2026Yanzhe Tang, Xinyu Shao, Yuxuan Hu +6Language-Conditioned Robot Manipulation Vision-Language Grounding
1Tsinghua University, Shenzhen, China. · 2Huawei Technologies Co., Ltd., Shenzhen, China. · 3National University of Singapore, Singapore.
iFLYTEK-Embodied-Omni Technical Report
Jun 24, 2026Yuan Zhang, Jingfei Ni, Guanchen Lu +12Unified Multimodal Models Robot Motion Generation
1iFLYTEK · University of Science and Technology of China · 2LindenBot
Reflective VLA: In-Context Action Consequences Make VLAs Generalize
Jun 23, 2026Qing Lian, Kent Yu, Lei ZhangICL for Robotics Generalization in Robotic Manipulation
Futian Laboratory · International Digital Economy Academy (IDEA) · Visincept
InSight: Self-Guided Skill Acquisition via Steerable VLAs
Jun 23, 2026Maggie Wang, Lars Osterberg, Stephen Tian +5Robot Skill Learning Large Language Model-Based Robot Planning
Stanford University · Princeton University
MANGO: Automated Multi-Agent Test Oracle Generation for Vision-Language-Action Models
Jun 23, 2026Pablo Valle, Shaukat Ali, Aitor Arrieta +1Automated Software Testing Vision-Language-Action Models
Mondragon University, Guipuzcoa, Spain · Simula Research Laboratory, Oslo, Norway · University of Ottawa, Ottawa, ON K1N 6N5, Canada +1
G $^3$ VLA: Geometric Inductive Bias for Vision-Language-Action Models
Jun 23, 2026Yue Peng, Yongzhe Zhao, Artur Habuda +5Multi-View Learning Robotic Manipulation
New York University Shanghai · Technical University of Denmark · MBZUAI - Mohamed bin Zayed University of Artificial Intelligence
Supervise What Survives: Geometry-Guided VLA Adaptation from Synthetic Robot Videos
Jun 23, 2026Danze Chen, Yanzhe Chen, Qiming Huang +3Visuomotor Policy Learning Vision-Language-Action Models
Show Lab, National University of Singapore
DriveStack-VLA: Render-Teacher Alignment for BEV-Based DeepStack Vision-Language-Action Model
Jun 23, 2026Jingke Wang, Zhenru Zhao, Shuangming Lei +8VLMs for Autonomous Driving End-to-End Autonomous Driving
Zhejiang University, Hangzhou, China. · The 2012 Labs, Huawei.
LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
Jun 22, 2026Rongxu Cui, Zongzheng Zhang, Jingrui Pang +11Robot Safety Vision-Language-Action Models
Institute for AI Industry Research (AIR), Tsinghua University · Beijing Academy of Artificial Intelligence (BAAI) · Beihang University +2
Flatness Preserves Instruction Following in Vision-Language-Action Models
Jun 22, 2026Haochen Zhang, Yonatan BiskSharpness-Aware Minimization Vision-Language-Action Models
Carnegie Mellon University
dVLA-RL: Reinforcement Learning over Denoising Trajectories for Discrete Diffusion Vision-Language-Action Models
Jun 22, 2026Yuhao Wu, Yitian Liu, Weijie Shen +13Vision-Language-Action Models VLMs for Robotics
ScaleLab at SJTU · 1Shanghai Jiao Tong University · 3Baidu AI Cloud +3
RECALL: Recovery Experience Collection for Active Lifelong Learning in Vision-Language-Action Models
Jun 22, 2026Ulas Berk Karli, Tesca FitzgeraldContinual Robot Learning Robot Skill Learning
Department of Computer Science Yale University United States
A Watermark for Vision-Language-Action and World Action Models
Jun 22, 2026Yule Liu, Shuai Liu, Jiaheng Wei +1Language Model Fingerprinting Vision-Language-Action Models
1Hong Kong University of Science and Technology (Guangzhou) · 3Xi’an Jiao Tong University · 2Wuhan University
BiliVLA: Scene-Aware Vision-Language-Action Model with Reinforcement Learning for Autonomous Biliary Endoscopic Navigation
Jun 22, 2026Jinsong Lin, Chi Kit Ng, Zhiyong Xiong +8Endoscopy Visuomotor Policy Learning
The Chinese University of Hong Kong, Hong Kong SAR, China. · The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. · University of Cambridge, Cambridge, United Kingdom. +2
Assistron: Bayesian Shared Autonomy with Off-the-shelf Vision-Language-Action Models
Jun 22, 2026Pinhao Song, Ze Fu, Yutong Hu +1Language-Conditioned Robot Manipulation Contact-Rich Robotic Manipulation
KU Leuven, Dept. Mechanical Engineering, Research unit Robotics, Automation and Mechatronics · KU Leuven, Dept. Electrical Engineering, Research unit Processing Speech and Images
Attacking the Trusted Imagination: Oracle-Level Integrity Attacks on Imagine-then-Act World Models
Jun 22, 2026Linghan Chen, Kaiyan Ji, Minyu GuoAdversarial Attacks Adversarial Attacks on VLA Models
Adelaide University
HiL-ResRL: A Model-Agnostic Finetuning Adapter via Human-in-the-loop Residual Reinforcement Learning
Jun 22, 2026Jingyi Liu, Zhaohong Mai, ShunSen He +5RL for Robotics Vision-Language-Action Models
1CloudRobo Lab, Huawei Cloud Computing Technologies Co., Ltd. · 2South China University of Technology.
Cloak: Zero-Shot Cross-Embodiment Manipulation by Masking the End-Effector from the VLA
Jun 22, 2026Michael Piseno, Guy Tevet, C. Karen LiuZero-Shot Robotic Manipulation Language-Conditioned Robot Manipulation
Stanford University
UniFS: Unified Fast-to-Slow Hierarchical Architecture for Vision-Language-Action Models
Jun 22, 2026Lin Sun, Zhiwei Guan, Conglin Wang +7Efficient VLM Inference Efficient VLA Model Inference
JD.com, China · Yiwu Research Institute of Fudan University, China · Tianjin University, China
PolicyTrim: Boosting Intrinsic Policy Efficiency of Vision-Language-Action Models
Jun 21, 2026Xianghui Wang, Feng Chen, Wenbo Zhang +4Efficient VLA Models Vision-Language-Action Models
1Sichuan University · 2Adelaide University · 3Beijing Institute of Technology
OpenHLM: An Empirical Recipe for Whole-Body Humanoid Loco-Manipulation
Jun 20, 2026Yingdong Hu, Haodong Zhu, Boyuan Zheng +6Humanoid Loco-Manipulation Long-Horizon Robotic Manipulation
Tsinghua University · Shanghai Qi Zhi Institute · Spirit AI
Wh0: Generative World Models as Scalable Sources of Egocentric Human Hand Manipulation Data
Jun 20, 2026Yangtao Chen, Zixuan Chen, Peiyang Wang +4HOI Generation Generalization in Robotic Manipulation
Shanghai Innovation Institute · Nanjing University · Shanghai Jiaotong University
A DVDrive Approach for doScenes Instructed Driving Challenge
Jun 19, 2026Zijian Fu, Xiangyang Chu, Mengshi Qi +3VLMs for Autonomous Driving Vision-Language-Action Models
Beijing University of Posts and Telecommunications · Xiaomi EV
UniviewVLA: A Unified Multiview Vision-Language-Action Model with World Modeling
Jun 19, 2026Tao Xu, Runhao Zhang, Zhijian Huang +7World Models for Robotics Multi-View Learning
1Tongji University · 2Shanghai Innovation Institute · 4Xiaomi EV +1
Decoupling the Declarative from the Procedural in Vision-Language-Action Models
Jun 19, 2026Nikolaos Tsagkas, Andreas Sochopoulos, Chris Xiaoxuan Lu +2Robotic Manipulation Transfer Learning
University of Edinburgh · UCL · Samsung AI Center - Cambridge
Semi-Supervised Vision-Language-Action Model
Jun 19, 2026Hongyang He, Jiuming Liu, Victor SanchezVLM Adaptation Vision-Language-Action Models
University of Warwick · University of Cambridge
ASCII Art Turns LLMs into VLA Controllers
Jun 19, 2026Yitao Jiang, Roy Xing, Luyang Zhao +3Language Model-Based Control AI Control
Dept. of Computer Science, Dartmouth College, Hanover, NH, USA. · Dept. of Electrical and Computer Engineering, Clemson University, Clemson, SC, USA. · Dept. of Mechanical and Aerospace Engineering, University of Houston, Houston, TX, USA.
VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models
Jun 19, 2026Florian Seligmann, Emiliyan Gospodinov, Enes Ulas Dincer +1Mahalanobis Distance Vision-Language-Action Models
Autonomous Learning Robots Lab, Karlsruhe Institute of Technology · FZI Forschungszentrum Informatik, Karlsruhe
NAC: Neural Action Codec for Vision-Language-Action Models
Jun 19, 2026Ahad Jawaid, Yu XiangNeural Audio Codecs Residual VQ
The University of Texas at Dallas
FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation
Jun 18, 2026Duc Minh Nguyen, Nghiem Tuong Diep, Binh Gia Nguyen +20VLM Adaptation Efficient VLA Models
VinRobotics, Vietnam · Center for AI Research, VinUniversity, Vietnam · University of Utah, USA +4

Jun 24, 2026J/K move · Enter open · S save

Decoupling Semantics and Geometric Grounding: Spatial Visual Prompts for Language-Conditioned Imitation Learning

Yanzhe Tang, Xinyu Shao, Yuxuan Hu, Siyu Chen+5

1Tsinghua University, Shenzhen, China. · 2Huawei Technologies Co., Ltd., Shenzhen, China. · 3National University of Singapore, Singapore.

Language-Conditioned Robot Manipulation Vision-Language Grounding Vision-Language-Action Models Robot Imitation Learning Visual Prompt Learning

Open paper AskPDF arXiv

While end-to-end Vision-Language-Action (VLA) models show promise in robotic manipulation, their monolithic paradigm inherently couples semantic reasoning and spatial control. This creates a severe alignment bottleneck, limiting precise target disambiguation in data-constrained imitation learning. To overcome this, we propose SVP-IL, a decoupled architecture that explicitly extracts spatial visual grounding from the action generation loop. By leveraging vision-language foundation models, we parse instructions into zero-shot geometric masks, translating language into explicit Spatial Visual Prompts (SVP). These priors are injected into a continuous action generator via a lightweight direct feature-level fusion mechanism. This integration provides explicit and uncorrupted spatial gradient guidance while ensuring highly stable optimization under low-data regimes. Extensive experiments demonstrate that SVP-IL significantly outperforms state-of-the-art VLAs and pure visuomotor baselines. Trained on as few as 50 to 100 demonstrations, SVP-IL improves average success rates on highly ambiguous language-conditioned tasks from 24.0% to 39.5%, achieving 67.8% on standard benchmarks. Real-world robotic experiments further validate its robustness and data efficiency in unstructured physical environments.