Vision-Language-Action (VLA)-based driving systems represent a significant paradigm shift in autonomous driving since, by combining traffic scene understanding, linguistic interpretation, and action generation, these systems enable more flexible, adaptive, and instruction-responsive driving behaviors. However, despite their growing adoption and potential to support socially responsible autonomous driving as well as understanding high-level human instructions, VLA-based driving systems may exhibit new types of hazardous behaviors. For instance, the integration of open-ended natural language inputs (e.g., user or navigation instructions) into the multimodal control loop may lead to unpredictable and unsafe behaviors that could endanger vehicle occupants and pedestrians. Hence, assuring the safety of these systems is crucial to help build trust in their operations. To support this, we propose a novel safety case design approach called RAISE. Our approach introduces novel patterns tailored to instruction-based driving systems such as VLA-based driving systems, an extension of Hazard Analysis and Risk Assessment (HARA) detailing safe scenarios and their outcomes, and a design technique to create the safety cases of VLA-based driving systems. A case study on SimLingo illustrates how our approach can be used to construct rigorous, evidence-based safety claims for this emerging class of autonomous driving systems.
ForesightSafety-VLA: A Unified Diagnostic Safety Benchmark for Vision-Language-Action Models
Brain-inspired Cognitive AI Lab, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所类脑认知智能实验室);Beijing Key Laboratory of Safe AI and Superalignment(北京市安全人工智能与超级对齐重点实验室);Beijing Institute of AI Safety and Governance(北京人工智能安全与治理研究所);Gaoling School of AI, Renmin University of China(中国人民大学高瓴人工智能学院);University of Chinese Academy of Sciences (UCAS)(中国科学院大学)
LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
Institute for AI Industry Research (AIR), Tsinghua University(清华大学人工智能产业研究院);Beijing Academy of Artificial Intelligence (BAAI)(北京智源人工智能研究院);Beihang University(北京航空航天大学);Eastern Institute of Technology, Ningbo(宁波东方理工大学);Shanghai Jiao Tong University(上海交通大学);Microsoft Research Asia (MSRA)(微软亚洲研究院)
SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning
Institute for Artificial Intelligence, Peking University(人工智能研究院,北京大学);PKU-PsiBot Joint Lab(北京大学PsiBot联合实验室);State Key Laboratory of General Artificial Intelligence, Peking University(通用人工智能国家重点实验室,北京大学);Zhongguancun Academy(中关村学院)
Concept-Based Dictionary Learning for Inference-Time Safety in Vision Language Action Models
Beijing Jiaotong University(北京交通大学);Provable Responsible AI and Data Analytics (PRADA) Lab(可证责任AI与数据分析实验室);King Abdullah University of Science and Technology(国王 Abdullah 科学技术大学);University of Auckland(奥克兰大学);RIKEN Center for Advanced Intelligence Project (AIP)(理化学研究所高级智能项目中心);Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)(穆罕默德·本·扎耶德人工智能大学)
SafeDojo: Safe Reinforcement Learning for VLA via Interactive World Model
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University(北京大学计算机学院多媒体信息处理国家重点实验室);Beijing Innovation Center of Humanoid Robotics(北京人形机器人创新中心);Nanyang Technological University(南洋理工大学);Hong Kong University of Science and Technology(香港科技大学);University of Electronic Science and Technology of China(电子科技大学)
VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Models
Institute for Artificial Intelligence, Peking University(北京大学人工智能研究院);Zhongguancun Academy(中关村学院);Beijing Academy of Artificial Intelligence(北京智源人工智能研究院);PKU-PsiBot Joint Lab(北京大学-PsiBot联合实验室)
VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer
Department of Automation, Tsinghua University(清华大学自动化系);Institute for Embodied Intelligence and Robotics, Tsinghua University(清华大学智能与机器人研究院);TetraBOT;DAMO Academy, Alibaba Group(阿里巴巴集团达摩院)
World-Env: Leveraging World Model as a Virtual Environment for VLA Post-Training
School of Computer Science and Engineering, Sun Yat-sen University, China(中山大学计算机科学与工程学院);AMap, Alibaba Group(阿里巴巴集团高德地图);Key Laboratory of Machine Intelligence and Advanced Computing, Ministry of Education, China(教育部机器智能与先进计算重点实验室)
Safe Learning for Contact-Rich Robot Tasks: A Survey from Classical Learning-Based Methods to Safe Foundation Models
Human-Robot Interfaces and Interaction Lab, Istituto Italiano di Tecnologia, Genova, Italy(人类-机器人接口与交互实验室,意大利技术研究院,热那亚,意大利);Ph.D. program of national interest in Robotics and Intelligent Machines (DRIM) and Università di Genova, Genoa, Italy(机器人与智能机器国家利益博士项目(DRIM)和热那亚大学,热那亚,意大利);Edwardson School of Industrial Engineering, Purdue University, West Lafayette, IN, USA(工业工程埃德华森学校,普渡大学,西拉法伊斯,美国)
Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Georgia Institute of Technology(佐治亚理工学院);University of Texas at Dallas(德克萨斯大学达拉斯分校)
Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future
HKUST(香港科技大学);Zhejiang University(浙江大学);National University of Singapore(新加坡国立大学);HKUST(GZ)(香港科技大学(广州));DAMO Academy, Alibaba(阿里巴巴达摩院);University of California, Los Angeles(加州大学洛杉矶分校);Xiaomi EV(小米电动车);Xi'an Jiaotong University(西安交通大学);Nanyang Technological University, Singapore(新加坡南洋理工大学)