智通财经APP获悉，小马智行-W(02026)再涨超3%，截至发稿，涨3.59%，报104港元，成交额1643.87万 港元。 消息面上，近日，小马智行-W披露2025年三季度业绩。美银证券指出，小马智行-W第三季收入按年增 长72%至2,540万美元，其中Robotaxi服务按年增长89%；毛利率达18.4%，按年及按季分别提升9.1及2.2 个百分点，因收入结构改善且Robotaxi服务贡献提高。期内研发费用为6,000万美元，按年增长80%，主 要投入于第七代车辆研发及研发人员扩充。非公认会计准则净亏损为5,500万美元，对比去年同期及上 季分别为4,100万美元及4,600万美元。 该行续指，截至11月23日，小马智行Robotaxi车队规模达961辆，管理层有信心在今年底突破1,000辆目 标，并预期2026年底将超过3,000辆。管理层又表示，车辆密度更高将缩短候车时间、提升用户体验， 从而令定价及利用率更高，并预期2026年Robotaxi收入增长将进一步加速，受稳健订单及营运效率提升 推动。 ...

Group 1 - The core point of the article highlights that Xiaoma Zhixing reported a significant increase in revenue and a notable loss in the third quarter of 2025 compared to the previous year [1][3] - In Q3 2025, the company achieved a revenue of $25.4 million, representing a 72% year-over-year growth, surpassing market expectations of $22.8 million [3] - The gross profit for Q3 2025 was reported at $4.67 million, indicating positive operational performance despite the losses [3] Group 2 - The non-GAAP basic and diluted loss per share for Q3 2025 was $0.14 (RMB 1.00), a significant decrease from $3.50 in Q3 2024 [1][3] - The net loss for Q3 2025 was $55 million, compared to a loss of $41.4 million in Q3 2024, reflecting ongoing challenges in profitability [3] - Xiaoma Zhixing announced that it expects to exceed its fleet expansion target, growing to over 3,000 vehicles by 2025 [1]

Core Insights - The article discusses the recent dissolution of a self-driving technology company, Haomo Technology, which has faced significant operational challenges and staff turnover [2][3]. Company Overview - Haomo Technology was established on November 29, 2019, as a subsidiary of Great Wall Motors, focusing on autonomous driving systems. The core team includes talents from Great Wall Motors and tech companies like Baidu and Huawei [6]. - The company previously made rapid advancements, launching its first last-mile delivery vehicle, "Xiao Mo Tuo," in November 2020, and introduced the MANA data intelligence system in December 2021, which has accumulated over 620,000 hours of learning time by 2023 [6][8]. Recent Developments - In 2023, Haomo Technology has seen a mass exodus of key personnel, including the departure of the chairman and several vice presidents, leading to concerns about its operational stability [5]. - The company's official communications have ceased since June 2023, with the last update being a holiday poster on October 1 [6]. Market Impact - Following the announcement of the company's closure, users of Haomo's products expressed concerns and dissatisfaction regarding their experience with the products [11].

Core Insights - The article discusses the advancements in autonomous driving through the introduction of the AD-R1 framework, which utilizes an Impartial World Model to address the "optimistic bias" found in traditional world models [2][3][57] - The framework allows for closed-loop reinforcement learning, enabling autonomous vehicles to learn from imagined failures, thereby improving safety and decision-making capabilities [9][57] Group 1: Background and Challenges - End-to-end autonomous driving has transformed the industry, but challenges remain, particularly with long-tail event failures due to distribution shifts [6] - Traditional reinforcement learning methods rely on external simulators, which have limitations such as simulation-to-reality gaps and lack of interactivity [6][9] - The need for a paradigm shift towards learning 3D/4D world models as high-fidelity generative simulators is emphasized [6] Group 2: Optimizing World Models - The AD-R1 framework introduces a new approach to mitigate the optimistic bias in world models, which often fail to predict negative outcomes [2][7] - The Impartial World Model (IWM) is designed to accurately reflect the consequences of both safe and unsafe behaviors, enhancing the reliability of predictions [3][10] - A counterfactual synthesis pipeline is implemented to generate a diverse training dataset that includes reasonable collision and lane deviation scenarios [3][10] Group 3: Experimental Results - The IWM significantly outperforms traditional models in risk prediction tasks, demonstrating its ability to accurately foresee failures [47][48] - The application of the AD-R1 framework leads to notable improvements in safety and performance metrics across various baseline models, with absolute increases in planning decision metrics (PDMS) of 1.7% and 1.1% [49] - Ablation studies reveal that the introduction of counterfactual synthesis and model-level optimizations are critical for enhancing causal fidelity and overall performance [51][52] Group 4: Future Directions - Future research may focus on generating counterfactual failure samples from unlabeled data to reduce reliance on high-precision annotations [57] - Expanding the framework to more complex multi-agent interaction scenarios could further enhance the robustness of autonomous driving systems in long-tail events [57]

Core Viewpoint - The article emphasizes the importance of end-to-end production in the automotive industry, highlighting the scarcity of qualified talent and the need for comprehensive training programs to address various challenges in this field [1][3]. Group 1: Course Overview - The course is designed to cover essential algorithms related to end-to-end production, including one-stage and two-stage frameworks, reinforcement learning applications, and trajectory optimization [3][9]. - It aims to provide practical experience and insights into production challenges, focusing on real-world applications and expert guidance [3][6]. Group 2: Course Structure - The course consists of eight chapters, each addressing different aspects of end-to-end production, such as task overview, algorithm frameworks, navigation information applications, and trajectory output optimization [9][10][11][12][13][14][15][16]. - The final chapter will share production experiences from various perspectives, including data, models, and strategies for system enhancement [16]. Group 3: Target Audience and Requirements - The course is aimed at advanced learners with a background in autonomous driving, reinforcement learning, and programming, although those with weaker foundations can still participate [17][18]. - Participants are required to have access to a GPU with recommended specifications and familiarity with relevant algorithms and programming languages [18].

11月26日，特斯拉全球副总裁陶琳在社交平台发文称，无论是美国、中国还是欧洲，Tesla全球各生产 基地的供应商选择都采用同样严格、客观的标准，完全基于质量、总成本、技术能力成熟度以及长期供 货连续性。供应商的原产国或地理来源不构成排除性标准。 | 2025年11月27日星期四 | NO.1 六部门：加快布局新领域新赛道，聚焦智能网联新能源汽车 11月26日消息，工业和信息化部等六部门印发《关于增强消费品供需适配性进一步促进消费的实施方 案》，其中提出，加快布局新领域新赛道。聚焦智能网联新能源汽车、智能家居、消费电子、现代纺 织、食品、绿色建材等重点行业，开展双百典型创新应用专项活动，打造百个标志性产品、百家创新企 业和一批可体验可推广的新产品首用场景样板。 点评：六部门发布的方案明确指出将加快布局智能网联新能源汽车等新领域，显示出国家对未来产业结 构的重视。这将刺激相关企业在新能源汽车、智能家居和绿色建材等领域的投资与创新，有助于提升长 远竞争力。 NO.2 特斯拉高管辟谣"供应链去中国化" 点评：北京市"十五五"规划强调了在"双碳"目标下推进新能源汽车发展的战略，这无疑将为相关企业提 供广阔的发展空间。随 ...

Core Viewpoint - The article discusses the recent financial performance of Li Auto and Pony.ai, highlighting their revenue, profitability, and strategic directions in the electric vehicle and autonomous driving sectors [3][8]. Group 1: Li Auto Financial Performance - Li Auto reported a revenue of 27.4 billion yuan for Q3, with an operating loss of 1.2 billion yuan and an operating profit of 3.4 billion yuan [4]. - The company provided a delivery guidance of 100,000 to 110,000 vehicles for Q4 [4]. - The main factor affecting Li Auto's profitability was the significant expenses related to the recall of the Li MEGA, which impacted Q3 results [14]. Group 2: Strategic Directions of Li Auto - Li Auto aims to return to a "startup management model" to enhance operational efficiency [6]. - The company is redefining its products as "embodied intelligent robots" rather than just vehicles or mobile devices, reflecting its focus on AI development [7]. - Li Auto plans to maintain a fully self-researched product line and is confident in achieving a historic sales breakthrough by 2026 [9][11]. Group 3: Market Outlook and Competition - The company anticipates that short-term policy changes, such as adjustments to purchase tax, may affect sales, but believes the market will eventually shift from policy-driven to product-driven sales [10]. - Li Auto is re-establishing its strategic position in the range-extended electric vehicle segment and will continue to offer family design and 5C ultra-fast charging technology across its lineup [12]. Group 4: Pony.ai Financial Performance - Pony.ai reported a revenue of 181 million yuan for Q3, marking a 72% year-on-year increase, with Robotaxi business revenue reaching 47.7 million yuan, up 89.5% [15]. - The company achieved single-vehicle profitability for its seventh-generation Robotaxi in Guangzhou, with an average of 23 orders per vehicle per day [15]. Group 5: Challenges and Future Prospects for Pony.ai - Despite achieving single-vehicle profitability, Pony.ai remains in an overall loss position, with a net loss of 392 million yuan for Q3 [17]. - The company is focusing on cost reduction through technology, operational efficiency, and light-asset expansion, moving closer to the goal of overall profitability in the Robotaxi sector [17].

小马智行 20251125 摘要 Pony.ai 通过次 IPO 筹集超过 8 亿美元，增强了资产负债表，为加速量 产和商业化提供资金保障，计划到年底扩展到 3,000 多辆机器人出租车， 并预计在 2026 年达到此规模。 公司推出第七代机器人出租车后，在广州等城市实现经济收支平衡，为 扩大车队规模创造了条件。第三季度机器人出租车收入同比增长 90%， 充电收入同比增长超过 200%。 Pony.ai 成为上海首家推出完全自动驾驶商业机器人出租车业务的公司， 并在深圳扩展商业运营区域，覆盖更大城市范围。 公司采用全栈集成技术，包括软件、硬件和运营，并通过高保真交互仿 真技术和 AI 学习评估器，不断改进驾驶标准，确保安全和效率。 Pony.ai 致力于全球市场扩张，已在中国、中东、东亚、欧洲和美国等 8 个国家建立业务，并计划通过与 Stellantis 联盟在欧洲部署测试车辆。 2025 年第三季度，公司收入达 4.05 亿美元，同比增长 72%，毛利率 提升至 18.4%。机器人出租车服务收入同比增长 89.5%，环比增长 338.7%。 公司计划利用香港上市所获资金巩固技术领先地位，增加研发投入，吸 ...

Group 1 - Major stock index futures are showing positive trends, with Dow futures up 0.18%, S&P 500 futures up 0.21%, and Nasdaq futures up 0.26% [1] - Chinese concept stocks are mixed, with Alibaba up 1.07%, Pinduoduo up 1.05%, and JD.com up 1.16%, while Xpeng Motors is down 2.26% and Bilibili is down 1.81% [1] - Dell's Q3 revenue increased by 11% year-over-year to $27.005 billion, driven by a surge in AI server orders totaling $12.3 billion, with a backlog of $18.4 billion [1] Group 2 - Uber is launching a fully autonomous Robotaxi service in Abu Dhabi, expanding its partnership with WeRide, although some routes will still have safety drivers [2] - Bank of America predicts Nvidia's market share will decline from 85% to 75% by 2030, despite the AI data center market expected to grow fivefold to approximately $1.2 trillion [3] - Japanese chipmaker Rapidus plans to build a 1.4nm wafer fab by FY2027, aiming to catch up with TSMC [4] Group 3 - Elon Musk announced that Tesla's Robotaxi fleet in Austin will double next month, although specific operational numbers have not been disclosed [4] - Li Auto reported Q3 revenue of 27.4 billion yuan, a decline of 36.2% year-over-year, and expects Q4 revenue guidance to be below market estimates [4] - Hedge fund manager Bill Ackman is seeking to raise $5 billion for a new closed-end fund, with $2 billion from institutional investors [5]

Group 1 - The core announcement from Youjia Innovation involves a proposed issuance of up to 14.01 million new H-shares at a price of HKD 14.88 per share, representing a discount of approximately 9.98% from the closing price of HKD 16.53 on November 25 [1] - If all shares are fully placed, the net proceeds are expected to be around HKD 204 million, with approximately 70% allocated for the development of L4 autonomous logistics vehicles and 30% for upgrading the foundational research platform [1] - The company aims to accelerate its technological reserves and commercialization resources through this new round of placement, which is expected to strengthen its competitive advantage and expand its market leadership in L4 autonomous driving business [1] Group 2 - Youjia Innovation has recently signed a strategic cooperation agreement with Xinjiao Automobile and Binghuo Di, aiming to design, produce, and deliver 800 units of competitive autonomous logistics vehicles [2] - This collaboration is based on the principle of "complementary advantages and win-win cooperation," focusing on the full-chain synergy of technology research and development, intelligent manufacturing, and large-scale operations [2] - The company’s self-developed Xiaozhu autonomous vehicle utilizes multi-sensor fusion and a five-fold safety redundancy architecture, enabling stable operation in various scenarios, with existing applications in high-dynamic environments like Shenzhen Huaqiangbei [2]