炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 来源：深网腾讯新闻 2025年12月5日，顶着"国产GPU第一股"光环的摩尔线程正式登陆科创板。上市首日，摩尔线程股价一 度涨超500%，股价拉升至688元/股，随后涨幅逐步收窄，回落至600元/股以下。截至上午10：38，摩尔 线程股价报575.01元/股。 摩尔线程本次在科创板共发行7000万股新股，发行价为114.28元/股。若以盘中最高价格688元/股计算， 中一签（500股）可赚约28.69万元。 在网下发行部分，共有267家机构投资者参与申购，总申购数量达704.06亿股，网下最终仅获配3920万 股。 就在普通投资者"一签难求"之际，一场资本狂欢拉开国产芯片的创富大幕。 张建中身价或超300亿，锁定期超3年 凭借早期入股或股权激励，摩尔线程至少有9位公司高层及核心技术人员因持股而在上市时身价过亿 元。根据发行方案，公司设立了员工持股平台，几位掌握核心技术与公司股权的创始人及高管，其个人 财富将实现跨越式增长。 摩尔线程曾在招股书中公布公司董事、审计委员会委员、高级管理人员及核心技术人员直接或间接持有 公司股份的情况。 摩尔 ...

机器之心报道 机器之心编辑部 一年一度的英伟达奖学金出炉了。 二十五年来，英伟达研究生奖学金计划（NVIDIA Graduate Fellowship Program）一直为研究生提供与英伟达技术相关的杰出工作支持。 今天，该计划宣布了 2026 年度的 10 位博士生获奖者，他们每人将获得最高 6 万美元的资助，以支持他们在涵盖计算创新所有领域的各项研究。 他们的研究工作聚焦于加速计算的前沿领域，包括了自主系统、计算机体系结构、计算机图形学、深度学习、编程系统、机器人技术和安全。 本年度的 10 位获奖者中有 8 位华人 。去年有 7 位华人博士生 入选，包括了上交、中科大、浙大校友。 接下来，我们一起了解下本年度获奖者的信息。 Jiageng Mao 南加州大学，获奖理由：利用来自互联网规模数据的各种先验知识解决复杂的物理人工智能问题，从而为现实世界中的具身智能体实现稳健、可推广的智能。 资料显示，Jiageng Mao 是南加州大学博士生，研究方向是物理人工智能，目标是通过开发机器人、计算机视觉和自然语言处理等领域的算法，将人工智能应用于 现实世界。据了解，他对直观物理学、大型视觉 - 语言（- 动作） ...

Financial Data and Key Metrics Changes - Consolidated revenue increased by 4% to $323 million, driven by growth in the three segments: NRS, fintech, and net2phone, which together grew by 16% [9][10] - Consolidated gross profit rose by 10% to a record $118 million, resulting in a gross margin of 37% [9][10] - Adjusted EBITDA reached a record $37.9 million, with an adjusted EBITDA margin of 11.7%, reflecting a 26% year-over-year growth [10][17] - Earnings per share (EPS) increased by 31% to $0.89, while non-GAAP diluted EPS climbed by 32% to $0.94 [10] Business Line Data and Key Metrics Changes - NRS reported a 22% increase in recurring revenue, reaching $35 million, with income from operations up 35% to $9 million [12] - The fintech segment's income from operations nearly doubled, increasing by 97% to $6 million, and adjusted EBITDA rose by 87% to $7.5 million [13] - Net2phone's revenue increased by 10% on a net reported basis, with income from operations rising 94% to $2 million [15] Market Data and Key Metrics Changes - Boss Money's digital channel generated 84% of transactions, with revenue growth of 20% [12] - NRS's merchant services revenue grew by 38%, while sales fees increased by 30%, offsetting a 15% decline in advertising and data revenue [12] Company Strategy and Development Direction - The company is focusing on holistic solutions that combine multiple offerings tailored to customer needs, particularly in the net2phone segment [6][14] - There are plans to explore adjacent markets, including food service and international markets, with potential acquisitions to accelerate growth [20][21] Management's Comments on Operating Environment and Future Outlook - Management expressed optimism about long-term growth opportunities through innovation in both current and adjacent markets [4][12] - The company is monitoring the impact of new federal excise taxes on cash remittances and its effects on retail businesses [18] Other Important Information - The Delaware Supreme Court dismissed all claims against IDT in the Straight Path Class Action suit, positively impacting the company [6] - The company held $220 million in cash and equivalents as of October 31, 2025, a decrease from the previous quarter [16] Q&A Session Summary Question: Expectations for special dividends or accelerated buybacks - Management indicated they are not looking at large M&A opportunities currently but may consider small acquisitions for NRS [18] Question: Additions of net payment processing accounts - The additions are primarily from businesses that require a POS [19] Question: Adjacent markets for growth - Management mentioned several adjacent markets, including food service and various verticals within the U.S. [20] Question: Potential for expanding into other countries - There is interest in expanding internationally, with potential acquisitions being considered [21] Question: Comments on IDT Global's top-line revenue - Management highlighted the resilience of the IDT Global business, focusing on maximizing gross profit despite fluctuations in revenue [23]

Core Viewpoint - Investors are increasingly confident that Alphabet's chip business, particularly its Tensor Processing Unit (TPU), will become a significant growth engine for future revenue [1][6]. Group 1: TPU's Impact on Alphabet's Stock - The success of TPU has been a major reason for Alphabet's stock price soaring by 31% in the fourth quarter, ranking tenth among S&P 500 constituents [6]. - TPU is viewed internally as a core advantage that accelerates the growth of Alphabet's cloud computing business [6]. Group 2: Market Potential and Future Projections - The market is optimistic about Alphabet's potential to sell TPU to third parties, which could open a new revenue stream valued at nearly $1 trillion [6][8]. - Gil Luria from DA Davidson suggests that if Alphabet seriously pursues TPU sales, it could capture 20% of the AI market, potentially becoming a $900 billion business [8]. Group 3: Competitive Advantages of TPU - TPU is an application-specific integrated circuit (ASIC) designed for specific purposes, particularly to accelerate machine learning, offering lower costs compared to NVIDIA's chips [10]. - Mark Iong from Homestead Advisers highlights that while Alphabet may not control the entire market, its ASIC chips provide a unique advantage, making it a leading player in the market [11]. - Alphabet is recognized as the only company with a leading position across various AI domains, including Gemini, Google Cloud, and TPU, which provides it with significant competitive advantages [11].

【TechWeb】12月4日消息，据外媒报道，苹果公司本周一在官网宣布，机器学习和人工智能战略高级 副总裁约翰•詹南德雷亚将卸任，转任顾问，直到明年春季退休，苹果同时宣布知名AI研究员阿马尔•苏 布拉马尼亚加入，出任AI业务副总裁，将向软件工程高级副总裁克雷格・费德里希汇报工作，他将主 持多个关键领域的工作，包括苹果基础模型、机器学习研究及AI安全与评估。 虽然詹南德雷亚要在明年春季才会从苹果退休，但已经 卸任机器学习和人工智能战略高级副总裁的他，个人信 息已经从苹果官网的管理层中撤下。 虽然詹南德雷亚要在明年春季才会从苹果退休，但已经卸任机器学习和人工智能战略高级副总裁的他， 个人信息已经从苹果官网的管理层中撤下。 在本周一宣布将卸任和退休时，约翰•詹南德雷亚的信息还在苹果官网的领导层中，但目前领导层中已 经看不到他的信息。 在约翰•詹南德雷亚撤下之后，苹果公司官网公布的管理层，目前共有18人，分别CEO库克，COO萨比• 汗，克雷格・费德里希等8名高级副总裁，葛越等7名副总裁，还有苹果院士菲儿•席勒。 不过在苹果官网上，他们新任命的AI业务副总裁阿马尔•苏布拉马尼亚，尚未出现在管理层中，后续是 否会加入， ...

在全球跨境贸易迭代升级背景下，联邦快递中国区总裁许宝燕近日接受上海证券报记者专访时表示，针对"黑五""圣诞季"等消费旺季，联邦 快递在强化运力外，还通过打造一系列数智技术和服务，助力跨境电商发展。中国已经成为联邦快递国际业务发展最快、最具潜力的地区之 一。未来联邦快递将持续聚焦中国市场，重点强化物流网络、跨境电商、可持续发展等关键领域。 上证报中国证券网讯（记者 王文嫣）在全球跨境贸易迭代升级背景下，联邦快递中国区总裁许宝燕近日接受上海证券报记者专访时表示， 针对"黑五""圣诞季"等消费旺季，联邦快递在强化运力外，还通过打造一系列数智技术和服务，助力跨境电商发展。中国已经成为联邦快递 国际业务发展最快、最具潜力的地区之一。未来联邦快递将持续聚焦中国市场，重点强化物流网络、跨境电商、可持续发展等关键领域。 全面实现数智化 接住购物季跨境物流需求 岁末年初，全球迎来购物狂欢季，电子商务领域持续火热。亚太地区是全球最大的零售电商市场，中国更是连续12年位居全球零售电商规模 首位。 面对这一巨大市场，许宝燕表示，联邦快递的SenseAware ID先进传感技术，通过轻便小巧的传感设备，每两秒发送一次包裹位置数据，相 ...

Core Viewpoint - Bloomberg has played a crucial role in connecting China's financial market with the global market over the past 30 years, particularly in the bond market, enhancing transparency and efficiency through data and technology [2][3]. Group 1: Milestones in Bloomberg's Development in China - The inclusion of Chinese bonds in the Bloomberg Global Aggregate Index in 2018 marked a significant milestone, increasing the weight of RMB bonds in the index from approximately 6% to about 10%, making it the third-largest after USD and EUR bonds [3]. - Bloomberg has supported various connectivity mechanisms, becoming the first overseas electronic trading platform to support both "Bond Connect" and direct investment models in 2019, facilitating investor participation in China's financial market [4]. - Continuous collaboration with Chinese financial institutions has been a focus, helping them enhance their global capabilities through data and technology, exemplified by a recent strategic partnership with Guotai Junan, China's largest securities firm [4]. Group 2: Changes and Impacts of China's Bond Market Opening - The current phase of China's bond market opening is characterized by a shift from "channel-based" to "institutional" opening, enhancing predictability, convenience, and professionalism for global investors [5]. - As of August 2025, 1,170 foreign institutions from 80 countries have entered the Chinese bond market, holding approximately 4 trillion RMB, indicating a growing interest in RMB assets despite fluctuations due to global interest rates [5][6]. - Institutional improvements are enhancing market transparency, liquidity, and predictability, making the experience for foreign investors more aligned with international practices [6]. Group 3: International Investors' Perspectives - International investors are increasingly focused on market transparency, liquidity, and expectations regarding exchange rates and policies when considering investments in Chinese bonds [7]. - Bloomberg aids investors in understanding these factors through data and analytical tools, providing insights into the Chinese bond market and macro policies [7]. Group 4: Innovations by Bloomberg - Bloomberg has leveraged technology to enhance market transparency and efficiency, utilizing AI, machine learning, and natural language processing to help clients extract key information from vast data [9]. - The BQUANT quantitative solution integrates data, computational power, and analytical models, significantly reducing the time required for strategy development and backtesting [10]. - A recent innovation includes a RMB bond repurchase trading solution that allows global investors to use bonds held through "Bond Connect" as collateral for electronic trading, improving financing and liquidity management [10]. Group 5: Future Expectations - Looking ahead, Bloomberg anticipates a clear direction for the opening of China's financial market, with further improvements in market mechanisms and the continued internationalization of the RMB [11]. - The expectation is that more Chinese financial institutions will integrate into the global market, requiring high-quality data, timely information, and reliable trading solutions, which Bloomberg aims to provide [11].

Core Viewpoint - The article emphasizes the critical importance of environmental perception algorithms in ensuring the safety of autonomous driving, highlighting the need for skilled professionals in this field [5]. Group 1: Job Responsibilities - The role involves accurately detecting and locating all objects in the surrounding environment, such as roads, pedestrians, vehicles, and bicycles, to ensure safe driving [5]. - Responsibilities include processing data from machine vision and LiDAR for autonomous driving applications, achieving complex perception functions like multi-target tracking and semantic understanding [5]. Group 2: Qualifications - A solid mathematical foundation is required, particularly in geometry and statistics [5]. - Proficiency in machine learning and deep learning, along with practical experience in cutting-edge technologies, is essential [5]. - Experience in algorithms related to scene segmentation, object detection, recognition, and tracking based on vision or LiDAR is necessary [5]. - Strong engineering skills are required, with expertise in C/C++ and Python, as well as familiarity with at least one other programming language [5]. - Knowledge of 3D imaging principles and methods, such as stereo and structured light, is important [5]. - A deep understanding of computer architecture is needed to develop high-performance, real-time software [5]. - A passion for innovation and creating technology to solve real-world problems is encouraged [5].

Core Insights - The article discusses the introduction of Improved MeanFlow (iMF), an enhanced version of the original MeanFlow (MF), which addresses key issues related to training stability, guidance flexibility, and architectural efficiency [1][4]. Model Performance - iMF significantly improves model performance by reformulating the training objective to a more stable instantaneous velocity loss and introducing flexible classifier-free guidance (CFG) [2][12]. - In the ImageNet 256x256 benchmark, the iMF-XL/2 model achieved a FID score of 1.72 in 1-NFE (single-step function evaluation), representing a 50% improvement over the original MF [2][18]. Model Configuration and Efficiency - The configurations of both MF and iMF models are detailed, showing a reduction in parameters and improved performance metrics for iMF models compared to MF models [3][19]. - For instance, the iMF-B/2 model has 89 million parameters and a FID score of 3.39, while the MF-B/2 model has 131 million parameters and a FID score of 6.17 [3][19]. Training Methodology - iMF's core improvement lies in reconstructing the prediction function, transforming the training process into a standard regression problem, which enhances optimization stability [4][11]. - The training loss is now based on instantaneous velocity, allowing for a more stable and standard regression training process [10][11]. Guidance Flexibility - iMF introduces a flexible classifier-free guidance mechanism, allowing the guidance scale to be learned as a condition, thus enhancing the model's adaptability during inference [12][14]. - This flexibility enables the model to learn average velocity fields under varying guidance strengths, unlocking CFG's full potential [12]. Contextual Conditioning - The iMF architecture employs an efficient in-context conditioning mechanism, replacing the large adaLN-zero module with multiple learnable tokens for various conditions, improving efficiency and reducing parameter count [15][17]. - This adjustment allows iMF to handle multiple heterogeneous conditions more effectively, leading to a significant reduction in model size and increased design flexibility [17]. Experimental Results - iMF demonstrates exceptional performance on challenging benchmarks, with the iMF-XL/2 model achieving a FID of 1.72 in 1-NFE, showcasing its superiority over many pre-trained multi-step models [18][20]. - In 2-NFE evaluations, iMF further narrows the performance gap between single-step and multi-step diffusion models, achieving a FID of 1.54 [20].

Group 1 - The company is recruiting interns for positions in Shenzhen, Shanghai, and Beijing, requiring in-person internships with a minimum commitment of three days per week for at least three months [1] - The application deadline for submitting resumes is December 31, 2025 [1] - Interns with outstanding performance may have the opportunity for full-time employment after the internship [1] Group 2 - Responsibilities include data processing, analysis, and assisting researchers with quantitative investment projects [2] - Interns will also assist in the development and tracking of financial engineering strategy models [2] - Additional tasks may be assigned by the team [2] Group 3 - Basic requirements include being a master's or doctoral student in STEM fields or financial engineering, with a strong preference for exceptional fourth-year students [3] - Proficiency in programming languages such as Python and familiarity with SQL databases are essential [3] - Candidates should possess strong self-motivation, analytical skills, and effective communication abilities [3] Group 4 - Preferred qualifications include a solid foundation in financial markets, familiarity with key concepts in stocks, bonds, futures, indices, and funds [4] - A strong mathematical background, research project experience, and published academic papers in SCI or EI are advantageous [4] - Familiarity with financial terminals like Wind, Bloomberg, and Tianruan, as well as knowledge of machine learning and deep learning, is a plus [4] Group 5 - Interested candidates should submit their resumes in PDF format to the specified email address, following a specific naming convention for the email subject [5] - Resumes not adhering to the naming format will be treated as spam [5] - Qualified candidates will be contacted for written tests and interviews after the resume collection deadline [5]