Core Insights - Logistics technology is the core competitiveness for modern logistics enterprises and is crucial for implementing national strategies and enhancing the resilience and security of supply chains [1] - The integration of "artificial intelligence + logistics" is creating new models and value, with technologies like unmanned driving and unmanned warehouses being scaled [1] - The 2026 Logistics Technology Conference will be held from March 11 to 13 in Huzhou, aiming to gather industry wisdom and showcase cutting-edge achievements [1] Forum and Topics - The conference will introduce five new forums focusing on industry pain points and cutting-edge tracks, including forums on embodied robotics, logistics data, multimodal transport, handling equipment, and unloading technology [2] - These forums will address key dimensions such as equipment upgrades, technology empowerment, new scenario implementation, and global collaboration [2] Conference Structure - The conference will feature a "main forum + specialized forums" structure, with the main forum focusing on how black technology empowers the logistics industry [3] - Sixteen specialized forums will cover topics like smart warehousing, logistics robotics applications, digital supply chain ecosystems, and low-altitude logistics technology [3] Industry Engagement - The conference will gather representatives from logistics companies, manufacturing, trade, government, academia, and investment institutions to explore new industry trends [4] - A dedicated area for showcasing embodied robotics will highlight advancements in key technologies and applications, promoting technology transfer and resource alignment [4] Historical Context and Future Direction - The Logistics Technology Conference has witnessed the leap of China's logistics industry from automation to intelligence, greening, and internationalization over the past decade [5] - The event aims to explore the development patterns and future directions of intelligent logistics, contributing to high-quality development and the construction of a strong logistics nation [5]

Core Viewpoint - Nobikang (02635.HK), focused on AI vertical applications, made its debut on the Hong Kong Stock Exchange, with its stock price soaring 363.75% on the first day, setting a record for the largest first-day gain of a new stock in the Hong Kong market this year [1] Group 1: IPO Details - Nobikang's IPO involved a global offering of 3.7866 million H-shares, accounting for 10% of total shares post-issuance, with a share price set at 80 HKD, raising approximately 303 million HKD in total funds and a net amount of about 257 million HKD [2] - The company plans to allocate the net proceeds from the IPO as follows: approximately 40% for continued research on core technologies, 40% for building a research and development center and new headquarters, 10% for seeking strategic investment and acquisition opportunities, and 10% for working capital and general corporate purposes [2] Group 2: Company Overview - Founded in 2015, Nobikang specializes in the industrial application of advanced technologies such as AI and digital twins in sectors like transportation, energy, and urban governance, providing integrated hardware and software solutions based on AI industry models [2] - The company's self-developed NBK-INTARI AI platform enables intelligent monitoring, detection, and operation and maintenance for clients in transportation, energy, and urban governance [2] Group 3: Market Position and Financial Performance - According to Zhaoshang Consulting, Nobikang is the second-largest provider of AI + power detection and monitoring systems in China, with a market share of approximately 5.9% in the rail transit sector for 2024 [3] - In the AI + detection and monitoring solutions for the rail transit industry, Nobikang ranks third with a market share of about 1.8% [3] - The revenue projections for Nobikang from 2022 to the first half of 2025 are approximately 253 million RMB, 364 million RMB, 403 million RMB, and 232 million RMB, with corresponding profits of approximately 63.16 million RMB, 88.57 million RMB, 115 million RMB, and 40.08 million RMB [3]

Core Viewpoint - The article discusses how Guangzhou Fantuo Digital Creative Co., Ltd. is leveraging digital twin technology to address challenges in industrial smart transformation, particularly in production line upgrades and equipment maintenance [1][5]. Group 1: Digital Twin Technology - Digital twin technology is identified as a key technology in the transition to Industry 4.0, enabling the control, prediction, and optimization of real production processes through virtual models [3][7]. - Fantuo has developed a production line twin platform that collects critical data such as product qualification rates and material waste rates via the Internet of Things (IoT), allowing for automatic alerts when issues arise [3][7]. Group 2: Production Line Optimization - The platform allows factories to design and validate new production line schemes in a digital environment before implementing them in reality, significantly reducing time and trial-and-error costs [3][7]. - The technology also aids in predicting equipment failures by integrating data from equipment with AI systems, thus enabling proactive maintenance rather than reactive repairs [3][7]. Group 3: User Accessibility and Training - To address concerns about the complexity of digital twin technology, Fantuo has created a no-code platform that allows users to easily build their digital twin environments through simple drag-and-drop operations [4][8]. - The company is involved in training robots in virtual environments before deploying them in real factories, enhancing operational efficiency [4][8]. Group 4: International Support and Case Studies - Fantuo has provided solutions for companies exporting equipment to Southeast Asia, utilizing mixed reality (MR) technology for remote guidance on installation and maintenance [4][8]. - The company has accumulated numerous industry case studies, including projects in the photovoltaic industry, tobacco, pharmaceuticals, and precision instruments, showcasing the practical applications of digital twin technology [4][8].

Core Viewpoint - Nobikang, a company specializing in artificial intelligence and digital twin technology, successfully listed on the Hong Kong Stock Exchange, achieving a remarkable first-day increase of 363.75%, setting a record for new stock price increases in nearly a decade [1][3]. Company Overview - Nobikang Artificial Intelligence Technology (Chengdu) Co., Ltd. officially listed on December 23, with the stock code 2635.HK, marking its entry into the Hong Kong stock market [3]. - The company is recognized as a national-level specialized and innovative "little giant," focusing on providing intelligent detection and monitoring solutions for major infrastructure such as rail transit and smart cities [3]. - Since its establishment, Nobikang has continuously deepened its expertise, with core technologies and products widely applied in domestic rail transit, becoming a key driver of industry intelligence upgrades [3]. Financial Performance - The initial public offering (IPO) was priced at 80 Hong Kong dollars per share, with a subscription rate of 3% for investors [1]. - Each hand of shares consists of 50 shares, resulting in a profit of 14,550 Hong Kong dollars per hand, excluding transaction fees [1].

Core Insights - The article discusses a new system called CLM (CPU-offloaded Large-scale 3DGS training) developed by a research team from New York University, which allows for city-scale 3D reconstruction using a single consumer-grade GPU, specifically the RTX 4090, by offloading memory-intensive parameters to CPU memory [1][20]. Group 1: 3D Gaussian Splatting (3DGS) Challenges - 3DGS has become a significant technology in neural rendering due to its high-quality rendering and speed, but it faces scalability issues when applied to complex scenes like urban areas, primarily due to GPU memory limitations [2]. - A high-precision 3DGS model typically contains tens of millions to over a billion Gaussian points, with each point requiring substantial memory for parameters and gradients, making it difficult to train on a single GPU [2][3]. Group 2: CLM System Design - CLM is designed to address the GPU memory bottleneck by dynamically loading Gaussian parameters from CPU memory only when needed, rather than keeping all parameters in GPU memory [3][4]. - The system employs three key mechanisms: 1. **Attribute Segmentation**: Only "key attributes" necessary for visibility are stored in GPU memory, while the majority of parameters are offloaded to CPU memory [5][6]. 2. **Pre-rendering Visibility Culling**: CLM calculates visible Gaussian points before rendering, reducing unnecessary computations and memory usage on the GPU [7][8]. 3. **Efficient CPU Utilization**: CLM minimizes data transfer delays through micro-batching, caching, and intelligent scheduling, allowing the CPU to effectively assist in training without slowing down the process [10][12]. Group 3: Performance Results - The implementation of CLM on an RTX 4090 allowed for the training of 102.2 million Gaussian points, a 6.7-fold increase compared to the traditional method, which could only handle 15.3 million points [13][14]. - Despite communication overhead, CLM achieved a training throughput of 55% to 90% of the enhanced baseline on the RTX 4090, and up to 86% to 97% on the slower RTX 2080 Ti [16]. - The quality of reconstruction improved significantly, with the PSNR of the 102.2 million point model reaching 25.15 dB, compared to 23.93 dB for the 15.3 million point model [18]. Group 4: Broader Implications - CLM represents a cost-effective solution for large-scale 3D reconstruction, addressing deployment challenges without the need for multi-GPU setups, which is beneficial for both academic and industrial applications [20]. - The growing demand for efficient and low-cost 3D reconstruction tools in areas like digital twins and large-scale mapping makes CLM's approach particularly relevant [20].

Group 1 - The core viewpoint of the news is that Zhongda An's stock has shown significant fluctuations, with a year-to-date increase of 66.60% and a recent decline over the past 20 days of 8.27% [1] - As of December 23, Zhongda An's stock price is 16.31 yuan per share, with a market capitalization of 2.286 billion yuan [1] - The company has experienced a net outflow of main funds amounting to 360.44 thousand yuan, with large orders showing a buy of 445.11 thousand yuan and a sell of 758.28 thousand yuan [1] Group 2 - Zhongda An's main business includes project management services, primarily engineering supervision, with revenue contributions from various sectors such as power supervision (20.73%) and civil engineering supervision (19.72%) [1] - As of September 30, the number of shareholders decreased by 41.07% to 8,906, while the average circulating shares per person increased by 69.71% to 13,520 shares [2] - For the first nine months of 2025, Zhongda An reported a revenue of 469 million yuan, a year-on-year decrease of 5.03%, and a net profit attributable to shareholders of 980.2 thousand yuan, down 56.30% year-on-year [2] Group 3 - Since its A-share listing, Zhongda An has distributed a total of 30.3829 million yuan in dividends, with 1.3632 million yuan distributed over the past three years [3]

Core Viewpoint - The stock of Shanghai Xinda Electric Co., Ltd. has experienced fluctuations, with a year-to-date increase of 71.18% but a recent decline in the last 20 and 60 days, indicating potential volatility in investor sentiment and market performance [1][2]. Company Overview - Shanghai Xinda Electric Co., Ltd. was established on March 10, 1995, and listed on December 24, 2010. The company specializes in the research, development, manufacturing, and sales of industrial automation control products, including elevator control and drive products, IoT solutions, energy-saving products, and robotics [2]. - The revenue composition of the company includes: General control and drive products and systems (53.91%), Robotics products and systems (19.35%), Elevator control products and systems (18.30%), and Other products (8.44%) [2]. - The company is categorized under the machinery equipment sector, specifically in specialized equipment for building facilities, and is associated with concepts such as digital twins, JD.com, smart logistics, express delivery, and machine vision [2]. Financial Performance - For the period from January to September 2025, the company achieved a revenue of 2.654 billion yuan, representing a year-on-year growth of 17.20%. The net profit attributable to shareholders was 4.3462 million yuan, showing a significant increase of 110.26% [2]. - Since its A-share listing, the company has distributed a total of 479 million yuan in dividends, with no dividends paid in the last three years [3]. Shareholder Information - As of September 30, 2025, the number of shareholders increased to 48,000, with an average of 11,026 circulating shares per person, a decrease of 1.85% from the previous period [2]. - Among the top ten circulating shareholders, Penghua Carbon Neutral Theme Mixed A (016530) holds 14.3631 million shares, an increase of 5.8973 million shares from the previous period, while Huaxia CSI Robotics ETF (562500) holds 12.7796 million shares, an increase of 2.3863 million shares [3].

Core Viewpoint - Digital Zhengtong's stock price has experienced a decline of 15.66% year-to-date, with significant net outflows of capital and a decrease in shareholder numbers, indicating potential challenges in its financial performance and market perception [1][2]. Financial Performance - As of September 30, Digital Zhengtong reported a revenue of 521 million yuan, a year-on-year decrease of 25.36%, and a net profit attributable to shareholders of -52.2 million yuan, reflecting a substantial decline of 354.59% [2]. - Cumulative cash dividends since the company's A-share listing amount to 257 million yuan, with 56.22 million yuan distributed over the past three years [3]. Stock Market Activity - On December 23, Digital Zhengtong's stock fell by 2.01%, trading at 14.59 yuan per share, with a total market capitalization of 9.062 billion yuan [1]. - The stock has seen a trading volume of 73.99 million yuan, with a turnover rate of 0.98% [1]. - The main capital outflow was 12.22 million yuan, with large buy orders accounting for 12.01% and sell orders for 28.53% of the total trading volume [1]. Shareholder Structure - As of September 30, the number of shareholders decreased by 12.64% to 68,700, while the average number of circulating shares per person increased by 14.47% to 7,499 shares [2]. - Among the top ten circulating shareholders, notable changes include a 40% reduction in holdings by Jinying Technology Innovation Stock A and a 4.95% decrease in holdings by Southern CSI 1000 ETF [3].

Core Viewpoint - Nobikang AI Technology (Chengdu) Co., Ltd., known as the "first AI + industrial intelligent inspection stock" in Hong Kong, officially listed on the Hong Kong Stock Exchange on December 23, with its stock code being 2635.HK, experiencing an opening increase of over 300% [1][3]. Company Overview - Founded in 2015 in Chengdu, Nobikang focuses on core technologies in artificial intelligence and digital twins, providing intelligent detection and monitoring solutions for major infrastructure such as rail transit and smart cities [3]. - The company has become a key driver of industry intelligence upgrades, serving over 80% of railway bureaus in China and achieving significant growth in its AI + energy business, which became its largest revenue segment in 2023 [3][4]. Investment and Financing - Nobikang has undergone five rounds of financing from December 2019 to June 2024, with notable investors including Bojiang Capital and Chengyu Fund, which have provided not only financial support but also assistance in connecting research resources and industry channels [4][5]. - The company’s valuation reached 2.13 billion yuan after a D+ round investment led by Chengdu Jiaozi Capital in 2024 [5][6]. Market Expansion and Performance - The successful listing of Nobikang reflects the rapid expansion of the "Chengdu Legion" in the capital market, with six new companies listed or approved for listing in Chengdu this year, bringing the total number of listed companies in the region to 154, the highest in Central and Western China [7][8]. - The overall market capitalization of listed companies in Chengdu exceeds 2 trillion yuan, with 122 A-share companies reporting a total revenue of 434.055 billion yuan in the third quarter [8].

来源：上海证券报·中国证券网 上证报中国证券网讯（记者 何漪）12月23日，诺比侃在港交所挂牌上市。截至9点57分，诺比侃大涨319%，报335.2港元/股。 | W | 诺比侃(2635) | | | | | --- | --- | --- | --- | --- | | | 12-23 09:57:37 | | | | | | 335.200 额 1.7亿 股本 3787万 市盈"" 111 | | | H | | | 255.200 319.00% 換 2.35% 市值 127亿 市净" 17.03 | | | | | 关于日本 | 王目 周Ｋ 月Ｋ | 車名 | | | | 叠加 | 均价:238.627 | 雷口 | | | | 700.000 | | 775.00% 卖10 337.000 10 | | | | | | 卖9 336.800 | | | | | | 卖8 336.600 | | | | | | 卖7 336.400 | | | | | | 卖6 336.200 | | | | 80.000 | | 0.000/ 25 336,000 | | | | | | 卖4 335.800 | ...