Core Viewpoint - The company, Jiadu Technology, announced the disposal of part of its equity in its associate company, CloudWalk Technology, to lock in investment gains, resulting in a tax-adjusted investment loss of 16.29 million yuan, which represents 14.18% of the company's audited net profit attributable to shareholders for 2024 [2]. Group 1: Equity Disposal Details - Between May 13, 2025, and May 26, 2025, Jiadu Technology disposed of 7,664,200 shares of CloudWalk Technology for a transaction amount of 99.64 million yuan (excluding transaction taxes) [2]. - This transaction marks the fifth time Jiadu Technology has reduced its holdings in CloudWalk Technology since May 2023, with a total of over 46 million shares disposed of, amounting to 643 million yuan in cash [4]. - As of the end of Q1 2025, Jiadu Technology held 25,802,100 shares of CloudWalk Technology, representing a 2.48% stake, which will decrease further after this disposal [4]. Group 2: Company Performance and Market Context - CloudWalk Technology, which went public on the STAR Market on May 27, 2022, has faced continuous losses for eight years, with a reported loss of 696 million yuan in 2024 [4]. - In Q1 2025, CloudWalk Technology reported revenue of 37.23 million yuan, a year-on-year decline of 31.56%, and a net loss attributable to shareholders of 124 million yuan [4]. - In contrast, Jiadu Technology reported a revenue of 2.254 billion yuan in Q1 2025, reflecting a year-on-year growth of 106.26%, with a net profit of 120 million yuan [4].

Group 1 - Shenghong Technology has preliminarily determined the inquiry transfer price at 65.85 yuan per share, with 24 institutional investors participating and a total of 25.73 million shares to be transferred [1] - Foton Motor plans to establish a new energy technology company with EVE Energy, investing 500 million yuan to enhance its new energy heavy truck business and battery leasing solutions [1] - Jiadu Technology disposed of 7.6647 million shares of CloudWalk Technology, realizing a post-tax investment loss of 16.29 million yuan, which is 14.18% of the company's audited net profit for 2024 [1] Group 2 - Rongsheng Environmental's controlling shareholder and actual controller, Feng Ronghua, has been subjected to compulsory measures by the police, but this matter is unrelated to the company, which continues normal operations [2] - Chongqing Steel has terminated the absorption merger of its wholly-owned subsidiary, New Port Changlong, to leverage its independent operational advantages in logistics and supply chain services [3] - Ha Welding Huatong's products can be used in controllable nuclear fusion projects, but the revenue from this sector is currently minimal, and there are no significant changes in the company's operations [4] Group 3 - Xinhua Medical's controlling shareholder plans to increase its stake in the company by 100 million to 200 million yuan within the next 12 months [4] - Pingzhi Information signed a framework agreement for AI computing power technology services with China Telecom, totaling approximately 246 million yuan, which will not significantly impact the company's annual performance [5]

Core Insights - The article discusses the development and support of the multimodal large model industry in China, highlighting various policies and initiatives at both national and local levels aimed at enhancing AI capabilities and applications [1][4][11]. Policy Development Timeline - In 2023, local policies began to emerge, focusing on computational power to encourage the development of large model technology and innovative application scenarios, starting with Guangdong, Beijing, and Shanghai. By 2024, more regions are expected to introduce relevant policies aimed at improving administrative efficiency [1]. - By 2025, government work reports will emphasize the ongoing promotion of the "Artificial Intelligence +" initiative, with a focus on supporting the widespread application of large models [1]. National Policy Summary - The Chinese government has implemented several measures to support the AI industry, particularly multimodal large models, which are seen as crucial products within the AI sector. The State Council has identified embodied intelligence as a future industry, promoting the integration of digital technology with manufacturing and market advantages [4][5]. - Key national policies include the "Guidelines for the Development of Artificial Intelligence Industry" and the "Three-Year Action Plan for Data Elements," which aim to enhance data utilization and promote high-quality economic development through data-driven initiatives [11][13]. Local Policy Highlights - Various provinces have introduced specific policies to support the development of AI large models. For instance, Guangdong aims to develop a comprehensive technology system for large models with trillion-parameter capabilities, while Beijing targets the creation of 3-5 advanced, controllable foundational model products by the end of 2025 [13][15]. - Local initiatives also include the establishment of intelligent computing centers and the promotion of AI applications in various sectors, such as manufacturing, healthcare, and urban governance [13][14]. Key Development Directions - The article outlines that provinces like Guangdong, Beijing, and Shanghai have set ambitious goals for the development of large models, focusing on creating a robust ecosystem for AI innovation and application [15]. - The emphasis is on fostering collaboration between government, industry, and academia to drive advancements in AI technologies and their practical applications across different sectors [15].

Core Insights - The article discusses the development and comparison of multimodal large models, emphasizing the integration of visual and language components to enhance understanding and generation capabilities in AI systems [1][7]. Multimodal Model Types - The mainstream approach for visual and language multimodal models involves using pre-trained large language models and image encoders, connected through a feature alignment module to enable deeper question-answer reasoning [1]. - CLIP, developed by OpenAI, utilizes a contrastive learning method to connect image and text feature representations, allowing for zero-shot classification by calculating cosine similarity between text and image embeddings [2]. - Flamingo, introduced in 2022, combines visual and language components, enabling text generation based on visual and textual inputs, and includes various datasets for training [5]. - BLIP, proposed by Salesforce in 2022, aims to unify understanding and generation capabilities for visual language tasks, enhancing model performance through self-supervised learning and addressing complex tasks like image generation and visual question answering [7]. - LLaMA integrates a visual encoder (CLIP ViT-L/14) with a language decoder, utilizing generated data for instruction fine-tuning, ensuring that visual and language tokens exist in the same feature space [8].

| 证券代码：688327 证券简称：云从科技 公告编号：2025-031 | | | | | | | --- | --- | --- | --- | --- | --- | | 云从科技集团股份有限公司 | | | | | | | 本公司董事会及全体董事保证公告内容不存在任何虚假记载、误导性陈述或 | | | | | | | 者重大遗漏，并对其内容的真实性、准确性和完整性依法承担法律责任。 | | | | | | | 重要内容提示： | | | | | | | ? 本次会议是否有被否决议案：无 | | | | | | | 一、 会议召开和出席情况 | | | | | | | （一） 股东大会召开的时间：2025 年 5 月 21 日 | | | | | | | （二） 股东大会召开的地点：上海市浦东新区川和路 弄张江人工智能岛 | 55 | | 11 | | | | 号楼会议室 | | | | | | | （三） 出席会议的普通股股东、特别表决权股东、恢复表决权的优先股股东及 | | | | | | | 其持有表决权数量的情况： | | | | | | | 普通股股东人数 553 | | | | | | | ...

证券代码：688327 证券简称：云从科技 公告编号：2025-031 云从科技集团股份有限公司 2024年年度股东大会决议公告 本公司董事会及全体董事保证公告内容不存在任何虚假记载、误导性陈述或 者重大遗漏，并对其内容的真实性、准确性和完整性依法承担法律责任。 重要内容提示： 本次会议是否有被否决议案：无 一、 会议召开和出席情况 （一） 股东大会召开的时间：2025 年 5 月 21 日 （二） 股东大会召开的地点：上海市浦东新区川和路 55 弄张江人工智能岛 11 号楼会议室 （三） 出席会议的普通股股东、特别表决权股东、恢复表决权的优先股股东及 其持有表决权数量的情况： | 1、出席会议的股东和代理人人数 | 554 | | --- | --- | | 普通股股东人数 | 553 | | 特别表决权股东人数 | 1 | | 2、出席会议的股东所持有的表决权数量 | 1,252,555,307 | | 普通股股东所持有表决权数量 | 21,910,427 | | 特别表决权股东所持有表决权数量（每份特别表决权股份的表决权数 | 1,230,644,880 | | 量为：6） | | | 3、出席会议的股 ...

北京市东城区建国门内大街 26 号新闻大厦 7 层、8 层 电话：010-88004488/66090088 传真：010-66090016 邮编：100005 北京国枫律师事务所 关于云从科技集团股份有限公司 2024 年年度股东大会的 法律意见书 1 3．本所及经办律师依据《证券法》《证券法律业务管理办法》《证券法律业务执业 规则》等规定及本法律意见书出具日以前已经发生或者存在的事实，严格履行了法定职 责，遵循了勤勉尽责和诚实信用原则，进行了充分的核查验证，保证本法律意见所认定 的事实真实、准确、完整，所发表的结论性意见合法、准确，不存在虚假记载、误导性 陈述或者重大遗漏，并承担相应法律责任； 4．本法律意见书仅供贵公司本次会议之目的使用，不得用作任何其他用途。本所 律师同意将本法律意见书随贵公司本次会议决议一起予以公告。 国枫律股字[2025]A0267号 致：云从科技集团股份有限公司（贵公司） 北京国枫律师事务所（以下简称"本所"）接受贵公司的委托，指派律师出席并见 证贵公司 2024 年年度股东大会（以下简称"本次会议"）。 本所律师根据《中华人民共和国公司法》（以下简称"《公司法》"）、《中华人 ...

Core Insights - The article emphasizes the growing importance and application of multimodal large models in various industries, highlighting their clearer commercial monetization paths compared to language-only models [1] Multimodal Large Models Applications - Multimodal large models are categorized into 11 application scenarios, with the top five being digital humans, gaming, advertising, social media, and intelligent marketing, indicating a high maturity level and significant attention in these areas [1] - Digital humans leverage multimodal technology to enhance human-computer interaction through natural language processing, voice synthesis, and realistic visual presentation, improving user experience [2][5] - In gaming, multimodal large models enhance interaction by allowing characters to understand player commands and respond contextually, creating a more immersive experience [5] - The advertising industry benefits from multimodal technology by automating content creation, personalizing ad delivery, and enhancing user engagement, leading to a more efficient and intelligent advertising ecosystem [8][10] - Social media platforms are being transformed by multimodal large models, improving content creation, user recommendations, interaction experiences, community governance, and commercialization [11][12]

近年来，越来越多的国产大模型表现惊艳，并在多个行业领域提供智能支撑。 进入2025年，政府工作报告进一步提出持续推进"人工智能+"行动，将数字技术与制造优势、市场优势 更好结合起来，并将"支持大模型广泛应用"首次写进报告。 作为新一轮科技革命和产业变革的重要驱动力量，人工智能大模型如何将感知、认知、决策的核心技术 闭环运用于跨场景、跨行业的智慧解决方案，帮助全面提升生产质效？ 5月9日，由21世纪经济报道、21世纪资本研究院组织的21世纪卓越董事会系列调研活动来到云从科技， 实地感受技术研发趋势与多场景、多产业融合的实践案例。 参与调研的有国内大型商业银行、头部基金公司相关人士。 从2023年大模型与通用人工智能开启"元年"至今，AI应用的发展整体经历了"三浪潮"。 第一浪是人工智能单点技术应用，包括较早的人脸识别、车辆识别、ReID跨境追踪、智慧航显等；第 二浪是多点闭环解决行业关键需求，各AI公司使用大量技术投入解决某一特定场景问题，包括智慧金 融、智慧安防、智慧机场等；第三浪广阔机遇，则是技术平台化和标准化趋势下带来的入口改变、流量 重构，通用模型和行业垂直模型应用加速发展。 而据了解，伴随着DeepS ...

Group 1 - CloudWalk Technology (688327) closed at 13.33 yuan on May 16, 2025, down 0.74%, with a turnover rate of 2.04%, trading volume of 153,900 shares, and a transaction amount of 207 million yuan [1] - On May 16, 2025, the capital flow for CloudWalk Technology showed a net outflow of 14.11 million yuan from main funds, a net outflow of 3.43 million yuan from speculative funds, and a net inflow of 17.55 million yuan from retail investors [1][6] Group 2 - CloudWalk Technology will hold its 2024 annual general meeting on May 21, 2025, at 15:00, using a combination of on-site and online voting [2] - The company will provide a network voting reminder service to better serve small and medium investors, ensuring timely participation and voting [2] Group 3 - CloudWalk Technology announced the listing of 79,822,447 shares of restricted stock from its initial public offering, which will be tradable starting May 27, 2025 [3][4] - The total share capital after the IPO was 740,670,562 shares, and the newly listed restricted shares account for 7.69% of the total share capital [3][4] - The company’s total share capital increased to 1,038,548,520 shares after implementing a capital reserve conversion plan and completing the first vesting period of the 2023 restricted stock incentive plan [3][4]