小鹏汽车方面向记者透露，2024年3月，刘先明加入小鹏汽车，今年6月以小鹏世界基座模型负责人的身 份亮相；现为小鹏汽车自动驾驶中心组织负责人，全面负责中心业务和组织管理工作，向公司CEO何小 鹏汇报。 今年4月14日，小鹏汽车首次披露正在研发720亿参数的超大规模自动驾驶大模型，即"小鹏世界基座模 型"，未来将通过云端蒸馏小模型的方式将基模部署到车端。在即将举办的年度AI科技日上，小鹏汽车 将公布其在物理AI领域取得的技术进展，主要涉及小鹏自研的物理世界AI基座模型。 中证报中证网讯(龚梦泽熊永红)10月10日，就"小鹏汽车自动驾驶中心负责人变更，刘先明将接替李力 耘担任该职位"这一消息，小鹏汽车方面向中国证券报记者回应称"属实"。 公开信息显示，刘先明2016年博士毕业于伊利诺伊大学厄巴纳-香槟分校电气和计算机工程专业。他曾 在Facebook(现Meta)、通用汽车旗下的自动驾驶公司Cruise任职，从事机器学习与计算机视觉领域的前 沿研究工作。 ...

图片来 源：界面图库 智通财经记者 | 宋佳楠 小鹏汽车自动驾驶业务现高层调整。 公开资料显示，刘先明2016年博士毕业于伊利诺伊大学厄巴纳-香槟分校电气和计算机工程专业。曾在 Facebook（现Meta）、通用汽车旗下的自动驾驶公司Cruise任职，从事机器学习与计算机视觉领域的前 沿研究工作。2024年3月，他正式加入小鹏汽车，担任AI团队负责人，同年6月以世界基座模型负责人 的身份公开亮相。此轮调整前，他已直接向公司董事长兼CEO何小鹏汇报工作。 今年6月，在美国纳什维尔举行的全球计算机视觉顶会CVPR 2025上，刘先明作为小鹏汽车代表，讲述 了其世界基座模型的技术细节，展示了基座模型在真实城市环境复杂路面的控车能力。 暂时休息的李力耘为80后，本科毕业于清华大学电子工程系，2013年获得美国纽约大学计算机专业博士 学位。2019 年6月，李力耘加入小鹏汽车，担任自动驾驶高级总监、智能驾驶决策规划算法和模拟器方 向负责人，负责小鹏汽车高速和城市自主导航辅助驾驶系统NGP的研发。 2023年8月，在小鹏汽车自动驾驶副总裁吴新宙离职后，李力耘开始全面负责小鹏智驾方案的技术研发 与量产落地，并于次年10月 ...

新京报贝壳财经讯（记者张冰）10月10日，就"小鹏汽车自动驾驶中心负责人变更，刘先明将接替李力 耘担任该职位"这一消息，小鹏汽车方面向新京报贝壳财经回应"属实"。 公开信息显示，刘先明 2016 年博士毕业于伊利诺伊大学厄巴纳-香槟分校电气和计算机工程专业。他曾 在 Facebook（现 Meta）、通用汽车旗下的自动驾驶公司 Cruise 任职，从事机器学习与计算机视觉领域 的前沿研究工作。2024年3月，刘先明加入小鹏汽车，今年6月曾以小鹏世界基座模型负责人的身份亮 相。 今年4月14日，小鹏汽车首次披露正在研发720亿参数的超大规模自动驾驶大模型，即"小鹏世界基座模 型"，未来将通过云端蒸馏小模型的方式将基模部署到车端。在即将举办的年度AI科技日上，小鹏汽车 将公布其在物理AI领域取得的技术进展，主要涉及小鹏自研的物理世界AI基座模型。 编辑 杨娟娟 校对 赵琳 ...

我们获悉，10 月 9 日，小鹏汽车自动驾驶中心发布内部信：李力耘不再担任自动驾驶中心负责 人， 世界基座模型负责人刘先明（Xianming Liu）接替该职位。 刘先明 2016 年博士毕业于伊利诺伊大学厄巴纳-香槟分校电气和计算机工程专业。他曾在 Facebook（现 Meta）、通用汽车旗下的自动驾驶公司 Cruise 任职，从事机器学习与计算机视 觉领域的前沿研究工作。 2024 年 3 月，刘先明加入小鹏汽车，担任小鹏 AI 团队负责人，今年 6 月曾以小鹏世界基座模 型负责人的身份亮相，在本次人事变动前已直接向小鹏汽车董事长兼 CEO 何小鹏汇报。据接 近小鹏汽车的人士透露，何小鹏近期频繁与刘先明就自动驾驶大模型展开深入交流。另一位知 情人士称，刘先明之前在小鹏从事的 AI 研究已经同时在为芯片、具身团队提供必要软件支 持。 目前李力耘仍在职。 AI 大模型是小鹏智驾团队技术研发的重点领域。 文 丨 李安琪 赵宇 编辑 丨 龚方毅 我们就上述事实性信息向小鹏汽车官方求证，小鹏方面回应称，自动驾驶 VLA 大模型是世界基座模 型的一个重要应用，但更难的是让世界基座模型具备对整个世界进行推演的能力，而 ...

Group 1 - The article emphasizes the growing interest in large model technologies, particularly in areas such as RAG (Retrieval-Augmented Generation), AI Agents, multimodal large models (pre-training, fine-tuning, reinforcement learning), and optimization for deployment and inference [1] - A community named "Large Model Heart Tech" is being established to focus on these technologies and aims to become the largest domestic community for large model technology [1] - The community is also creating a knowledge platform to provide industry and academic information, as well as to cultivate talent in the field of large models [1]

Group 1 - The core point of the article is that Chasing Technology has officially announced its entry into the automotive industry, planning to launch a luxury electric vehicle that will compete with Bugatti Veyron by 2027 [2][4] - Chasing Technology aims to establish itself as a new benchmark in the global ultra-high-end electric vehicle market, focusing on high performance, intelligence, and luxury [2][4] - The company has formed a nearly 1,000-person team dedicated to vehicle manufacturing and is expanding its capabilities in key technology areas such as high-speed digital motors and AI algorithms [4][5] Group 2 - As of the end of 2024, Chasing Technology has applied for a total of 6,379 patents globally, with 45% being invention patents covering core areas of intelligent automotive technology [5] - The company has a global presence, operating in over 100 countries and regions, with more than 6,000 offline stores and serving over 30 million households [5] - Chasing Technology has established a subsidiary, "Starry Sky Plan (Shanghai) Automotive Technology Co., Ltd.," with a registered capital of 1 billion yuan, and is planning a factory in Shanghai [7] Group 3 - The company is actively recruiting for positions related to autonomous driving technology, indicating a focus on high-level autonomous driving solutions [6][7] - Chasing Technology has already secured some intention contracts with first-tier automotive companies for its autonomous driving technology [6] - The company has raised funds through six rounds of financing, with notable investors including Shunwei Capital and Xiaomi Group [7][8] Group 4 - Chasing Technology is not the only home appliance company venturing into the automotive sector, as other companies like Midea, Gree, and Haier have also made similar moves [12][13] - The company has diversified its business by announcing new ventures in home appliances and drones in 2025, showcasing its ambition to expand beyond its original market [9][10][11]

Core Viewpoint - The article discusses the advancements of the Li Auto VLA driver model, highlighting its enhanced capabilities in understanding semantics, reasoning, and trajectory planning, which are crucial for autonomous driving [1][3]. Summary by Sections VLA Model Capabilities - The VLA model has improved in three main areas: better semantic understanding through multimodal input, enhanced reasoning abilities via thinking chains, and closer alignment with human driving intuition through trajectory planning [1]. - Four core capabilities of the VLA model are showcased: spatial understanding, reasoning, communication and memory, and behavioral capabilities [1][3]. Development and Research Trends - The VLA model has evolved from VLM+E2E, incorporating various cutting-edge technologies such as end-to-end learning, trajectory prediction, visual language models, and reinforcement learning [5]. - While traditional perception and planning tasks are still being optimized in the industry, the academic community is increasingly shifting focus towards large models and VLA, indicating a wealth of subfields still open for research [5]. VLA Research Guidance Program - A VLA research paper guidance program has been initiated, receiving positive feedback, with many students eager for a second session. The program aims to help participants systematically grasp key theoretical knowledge and develop their own research ideas [6]. - The program includes a structured curriculum over 14 weeks, covering topics from traditional end-to-end autonomous driving to writing methodologies for research papers [9][11]. Enrollment and Course Structure - The program is limited to 6-8 participants per session, targeting students at various academic levels interested in VLA and autonomous driving [12]. - Participants will gain insights into classic and cutting-edge papers, coding implementations, and methods for selecting research topics and writing papers [13][14]. Course Highlights - The course emphasizes a comprehensive learning experience with a "2+1" teaching model, involving main instructors and experienced research assistants to support students throughout the program [22]. - Students will receive guidance on coding, research ideas, and writing methodologies, culminating in the production of a research paper draft [31][32]. Required Skills and Resources - Participants are expected to have a foundational understanding of deep learning, basic programming skills in Python, and familiarity with PyTorch [19]. - The program encourages the use of high-performance computing resources, ideally with multiple GPUs, to facilitate research and experimentation [19]. Conclusion - The VLA model represents a significant advancement in autonomous driving technology, with ongoing research and educational initiatives aimed at fostering innovation in this field [1][5][31].

Group 1 - The 2025 World Artificial Intelligence Conference opened in Shanghai, featuring a forum on AI-enabled innovation in autonomous driving, where new licenses for intelligent connected vehicles were issued [1][2] - The new licenses cover both intelligent taxi and freight operations, with four vehicle models receiving approval [1] - SAIC Group became the only company in the industry to obtain dual licenses for both passenger and commercial vehicles [1][3] Group 2 - SAIC Zhiji Automobile, in collaboration with Xiangdao Mobility and Saike Intelligent, will launch a trial operation of an L4-level Robotaxi service from Shanghai International Tourism Resort to Pudong International Airport, marking a first in the country for high-speed road access to an airport [3] - SAIC Zhiji Robotaxi has accumulated over 300,000 orders and has completed millions of kilometers in autonomous driving [4] - SAIC Youdao Zhitu will pioneer L4-level autonomous driving commercial operations in the domestic freight sector, with over 500,000 standard containers already transported at Yangshan Port [4] Group 3 - Pony.ai has established a commercial network for autonomous driving across four major cities in China, including Beijing, Shanghai, Guangzhou, and Shenzhen [4][5] - WeRide, in partnership with Chery Automobile and Jinjiang Taxi, has received approval for a driverless public road passenger service in Shanghai, expanding its Robotaxi operations to a global tenth city [5] Group 4 - The "Mosu Zhixing" action plan was released during the forum, aiming to establish Shanghai as a global leader in high-level autonomous driving by 2027, with a target of 6 million passenger rides and 800,000 container transports [7] - The plan includes the establishment of a digital twin training ground and the opening of 2,000 square kilometers of operational area, with over 5,000 kilometers of roads [8] - By 2024, Shanghai's automotive industry is projected to reach a production value of 703.5 billion yuan, with 1.645 million new energy vehicles promoted, maintaining its position as a global leader [8]

Core Viewpoint - Xpeng Motors is advancing its self-research capabilities with a clear roadmap for self-developed chips, robotics, intelligent driving, and foundational models, expecting strong sales and revenue growth in the coming years [1][2]. Group 1: Product Launch and Sales Expectations - Xpeng Motors has launched the new X9 electric MPV, starting at a price of 359,800 yuan, with expectations for sales to surpass previous models [1][5]. - The X9 is available in four versions, all equipped with Turing AI driving technology and featuring zero-gravity seats in the rear [5]. - The company anticipates a strong product cycle with the upcoming G7 and extended-range models later this year [1]. Group 2: AI and Autonomous Driving Developments - Xpeng has revealed its development of a 720 billion parameter large-scale autonomous driving model, named the "Xpeng World Base Model," which is set to support advanced driving capabilities [2][4]. - The company has established a computing cluster to support the training of this model, achieving a computing power of 10 EFLOPS with high operational efficiency [3]. - The model is designed to perform complex reasoning and interaction with the physical world, enhancing the capabilities of future intelligent driving systems [4]. Group 3: Chip Development and Future Technologies - The Turing AI chip, designed for high-performance AI applications, is expected to begin mass production in the second quarter of 2025, providing significant computational power for autonomous driving and robotics [6]. - Xpeng's plans include the mass production of flying cars by 2026 and the industrial-scale production of the IRON robot, which will utilize the Turing AI chip [6].

第一财经每日精选最热门大公司动态。 【今日推荐】 问界就"沈阳工业展览馆事故"发声明：不满足AEB触发条件 问界汽车用户服务部发布《关于辽宁沈阳工业展览馆M7Pro事故的声明》称，4月13日，一辆问界 M7Pro在辽宁沈阳工业展览馆发生碰撞事故，造成一名人员受伤。问界汽车深表歉意，已积极协助 就医，经检查伤者有轻微挫伤，当日已出院。问界称，经初步调查，事故发生于车展撤展前夕，车辆 未处于展车模式，导致车辆被误启动。因主驾未系安全带，且方向盘大角度干预，不满足AEB触发 条件。问界汽车表示，对此事故给各方带来的困扰，再次深表歉意，问界汽车将持续与车展主办方等 相关方积极协商、妥善处理后续事项。 苹果二手机回收已涨500元 4月14日，杭州手机数码市场一商户表示，苹果二手机回收一天一个价，目前涨400-500元。苹果 16promax 256G回收价原先是7300元，现在最高7800元。受关税影响，大家出现持货观望情绪， 都不愿意出手。（中国蓝新闻） 宇树科技：美国加关税并未有太大影响，没有哪家企业不做美国市场就不行了 4月14日，宇树科技华南区大区总监林志龙在接受采访时表示，美国加关税并未有太大影响，国外是 很 ...