Vision-Language-Action (VLA) models leverage Multimodal Large Language Models (MLLMs) for robotic control, but recent studies reveal that MLLMs exhibit limited spatial intelligence due to training predominantly on 2D data, resulting in inadequate 3D perception for manipulation tasks. While recent approaches incorporate specialized 3D vision models such as VGGT to enhance spatial understanding, they employ diverse integration mechanisms without systematic investigation, leaving the optimal fusion strategy unclear. We conduct a comprehensive pilot study comparing nine VGGT integration schemes on standardized benchmarks and find that semantic-conditioned gated fusion, which adaptively balances 2D semantic and 3D geometric features based on task context, achieved the strongest performance among all nine evaluated fusion schemes in our pilot study. We present 3D-Mix, a plug-and-play module that integrates into diverse VLA architectures (GR00T-style and $π$-style) without modifying existing MLLM or action expert components. Experiments across six MLLM series (nine model variants, 2B--8B parameters) on SIMPLER and LIBERO show that 3D-Mix delivers consistent performance gains, averaging +7.0% on the out-of-domain (OOD) SIMPLER benchmark across all nine GR00T-style variants, establishing a principled approach for enhancing spatial intelligence in VLA systems.
MixVLA: Adaptive Mixing of Non-Invariant Information for Generalizable Vision-Language-Action Models
Fudan University（复旦大学）；Shanghai AI Laboratory（上海人工智能实验室）；Harbin Institute of Technology, Shenzhen, China（哈尔滨工业大学（深圳））；Northwestern Polytechnical University（西北工业大学）；TeleAI, China Telecom Corp Ltd（中国电信股份有限公司TeleAI）
TORL-VLA: Tactile Guided Online Reinforcement Learning for Contact-Rich Manipulation
Meituan（美团）；Beijing Institute of Technology（北京理工大学）；Beihang University（北京航空航天大学）；State Key Lab of Multimodal Artificial Intelligence Systems, Institute of Automation, CAS（中国科学院自动化研究所多模态人工智能系统国家重点实验室）；China University of Mining and Technology (Beijing)（中国矿业大学（北京））
PAMAE: Phase-Aware-MoE Action Experts Towards Reliable Flow-Matching Vision-Language-Action Policies
Department of Artificial Intelligence, School of Informatics, Xiamen University（厦门大学信息学院人工智能系）；Department of Computer Science, Aberystwyth University（阿伯里斯特威斯大学计算机科学系）
PLaW-VLA: Predictive Latent World Modeling for Vision-Language-Action Policies
Jilin University（吉林大学）；Astribot；Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））；The Hong Kong Polytechnic University（香港理工大学）；The University of Tokyo（东京大学）
MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation
Nanjing University（南京大学）；The Hong Kong Polytechnic University（香港理工大学）；Peking University（北京大学）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）；The Hong Kong University of Science and Technology（香港科技大学）
CosFly-VLA: A Spatially Aware Vision-Language-Action Model for UAV Tracking
Autel Robotics（大疆创新科技有限公司）；Northeast Normal University（东北师范大学）；Southern University of Science and Technology（南方科技大学）；Peking University（北京大学）；University of Hong Kong（香港大学）
Beyond Imitation: Reinforcement Learning-Based Sim-Real Co-Training for VLA Models
Tsinghua University（清华大学）；Harbin Institute of Technology（哈尔滨工业大学）；Peking University（北京大学）；Carnegie Mellon University（卡内基梅隆大学）；Shanghai AI Laboratory（上海人工智能实验室）；Zhongguancun Academy（中关村学院）
ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weighting
The University of Hong Kong（香港大学）；Institute of Artificial Intelligence (TeleAI), China Telecom（中国电信人工智能研究院（TeleAI））；Shenzhen Research Institute of Northwestern Polytechnical University（西北工业大学深圳研究院）
AIR-VLA+: Decoupling Movement and Manipulation via Cascaded Dual-Action Decoders with Asymmetric MoE for Aerial Robots
The Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Automation, Beijing Institute of Technology（北京理工大学自动化学院）；College of Automotive and Energy Engineering, Tongji University（同济大学汽车与能源工程学院）；School of Transportation Science and Engineering, Beihang University（北京航空航天大学交通科学与工程学院）；Information Science, East China Normal University（华东师范大学信息科学）
Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learning
MoE key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University（人工智能混合专家实验室，人工智能研究所，上海交通大学）；School of Automation and Intelligent Sensing, Shanghai Jiao Tong University（自动化与智能感知学院，上海交通大学）；School of Computer Science, Shanghai Jiao Tong University（计算机科学学院，上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院，清华大学）；The University of Hong Kong（香港大学）；Tongji University（同济大学）；D-Robotics；Key Laboratory of System Control and Information Processing, Ministry of Education of China（系统控制与信息处理重点实验室，中华人民共和国教育部）；Shanghai Key Laboratory of Integrated Administration Technologies for Information Security（上海信息安全管理集成技术重点实验室）
DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization
School of Data Science, The Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳）数据科学学院）；Shenzhen Loop Area Institute（深圳河套学院）；Zhejiang University（浙江大学）；Rutgers University-New Brunswick（罗格斯大学新布朗斯维尔回声分校）；Shanghai AI Laboratory（上海人工智能实验室）；Jiangxing Intelligence Technology Inc.（江行智能科技有限公司）
FedVLA: Federated Vision-Language-Action Learning with Dual Gating Mixture-of-Experts for Robotic Manipulation
National University of Defense Technology（国防科技大学）；Bytedance Seed（字节跳动 Seed）；Shenzhen MSU-BIT University（深圳北理莫斯科大学）；Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（广东人工智能与数字经济实验室（深圳））
The Moving Eye: Enhancing VLA Spatial Generalization via Hybrid Dynamic Data Collection
Lion Rock AI Lab, China Merchants Research Institute of Advanced Technology（狮岩人工智能实验室，中国商人先进科技研究院）；The Hong Kong University of Science and Technology（香港科学与技术大学）；Nankai University（南开大学）
Algorithm-Architecture Co-Design for Efficient VLA Inference via Speculative Inference and Verification
School of Computer Science, Shanghai Jiao Tong University（上海交通大学计算机科学学院）；Computer Science, King Abdullah University of Science and Technology（阿卜杜拉国王科技大学计算机科学学院）
DualVLA: Building a Generalizable Embodied Agent via Partial Decoupling of Reasoning and Action
MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition, University of Science and Technology of China（脑启发智能感知与认知国家重点实验室，中国科学技术大学）；State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University（多媒体信息处理国家重点实验室，北京大学计算机学院）；CUHK（香港大学）
OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism
Institute for AI Industry Research (AIR)（人工智能产业研究院）；Department of Electronic Engineering（电子工程系）；University of Science and Technology of China（中国科学技术大学）
Never Too Late for Force: Accelerating VLA Post-Training with Reactive Force Injection
Shanghai Jiao Tong University（上海交通大学）；Shanghai Innovation Institute（上海创新研究院）；Southern University of Science and Technology（南方科技大学）；Noematrix Ltd.（诺玛矩阵有限公司）
TacCoRL: Integrating Tactile Feedback into VLA via Simulation
University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；University of California, San Diego（加利福尼亚大学圣迭戈分校）；University of Electronic Science and Technology of China（电子科技大学）；Peking University（北京大学）；University of Utah（犹他大学）
VANE: Reliable Test-Time Training for Vision-Language-Action Models via Future Visual Representation Prediction
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Beijing University of Posts and Telecommunications（北京邮电大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Li Auto Inc.（理想汽车）
HMVLA: Hyperbolic Multimodal Fusion for Vision-Language-Action Models
Harbin Institute of Technology（哈尔滨工业大学）；Chongqing Research Institute of HIT（重庆HIT研究 institute）；Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（广东人工智能与数字经济实验室（深圳））
