拓尔思信息技术股份有限公司 简式权益变动报告书 上市公司名称：拓尔思信息技术股份有限公司 股票上市地点：深圳证券交易所 信息披露义务人之一：信科互动科技发展有限公司 通讯地址：西藏自治区拉萨市达孜区工业园区企业服务中心 2 楼 201-1 室 信息披露义务人之二：李渝勤 通讯地址：北京市海淀区建枫路（南延）6 号院 3 号楼 邮政编码：100089 联系电话：010-64848899 信息披露义务人之三：施水才 通讯地址：北京市海淀区建枫路（南延）6 号院 3 号楼 邮政编码：100089 联系电话：010-64848899 股份变动性质：股份减少（因大宗交易减持公司股份导致持有公司股份比例 减少） 简式权益变动报告书签署日期：2025 年 7 月 14 日 股票简称：拓尔思 股票代码：300229 邮政编码：850100 联系电话：0891-64843223 信息披露义务人声明 一、信息披露义务人依据《中华人民共和国证券法》（以下简称《证券法》）、 《上市公司收购管理办法》（以下简称《收购办法》）、《公开发行证券的公司信息 披露内容与格式准则第 15 号—权益变动报告书》（以下简称"准则 15 号"）及相 ...

拓尔思（300229）今日官微消息，公司为相关机构研究加密货币市场动态、稳定币运行机制、监管合规 策略等提供AI赋能的数据分析与挖掘能力。同时，融合自然语言处理(NLP)和知识图谱技术，赋能金融 机构快速洞察全球稳定币政策动向。另外，依托融合另类大数据的金融风控大模型与智能预警能力，公 司为涉及数字货币及区块链应用的业务提供实时风险监测与防范能力。(人民财讯) ...

Group 1 - The digital currency concept index rose by 2.31%, ranking 10th among concept sectors, with 78 stocks increasing in value [1][2] - Notable gainers included Gu Ao Technology with a 20% limit up, and Jin Zheng Co., Ji Da Zheng Yuan, and Heng Bao Co. also hitting the limit up [1] - The sector saw a net inflow of 3.174 billion yuan, with 50 stocks receiving net inflows, and 15 stocks exceeding 100 million yuan in net inflow [2] Group 2 - The top net inflow stocks included Heng Bao Co. with 470 million yuan, followed by Jin Zheng Co. and Yu Xin Technology with 453 million yuan and 435 million yuan respectively [2][3] - Gu Ao Technology, Ji Da Zheng Yuan, and Ge Er Software had the highest net inflow ratios at 33.76%, 21.20%, and 16.52% respectively [3] - The digital currency sector's performance was supported by significant trading volumes, with Heng Bao Co. and Jin Zheng Co. showing high turnover rates of 45.74% and 16.10% respectively [3][4]

Core Viewpoint - The MLOps concept has shown a significant increase of 3.05%, ranking second among concept sectors, with notable stocks like StarRing Technology, Yuxin Technology, and TuoerSi leading the gains [1][2]. Market Performance - The MLOps concept sector saw a net inflow of 899 million yuan, with 13 stocks experiencing net inflows, and 5 stocks exceeding 30 million yuan in net inflow. Yuxin Technology led with a net inflow of 435 million yuan, followed by TuoerSi and Runhe Software with net inflows of 188 million yuan and 183 million yuan respectively [2][3]. Stock Performance - Key stocks in the MLOps sector include: - Yuxin Technology: Increased by 7.18% with a turnover rate of 16.64% and a net inflow of 435 million yuan, resulting in a net inflow ratio of 12.38% [3] - TuoerSi: Increased by 6.72% with a turnover rate of 11.30% and a net inflow of 188 million yuan, resulting in a net inflow ratio of 10.11% [3] - Runhe Software: Increased by 2.77% with a turnover rate of 5.61% and a net inflow of 183 million yuan, resulting in a net inflow ratio of 8.31% [3] Additional Insights - Other stocks with notable performance in the MLOps sector include: - Dongfang Guoxin: Increased by 4.00% with a net inflow of 54.9 million yuan and a net inflow ratio of 6.57% [3] - StarRing Technology: Increased by 12.40% with a net inflow of 24.77 million yuan and a net inflow ratio of 5.20% [3]

Market Overview - The Shanghai Composite Index closed at 3546.50 points, above the annual line, with a gain of 1.05% [1] - The total trading volume of A-shares reached 10310.63 billion yuan [1] Stocks Breaking Annual Line - A total of 55 A-shares have broken above the annual line today, with notable stocks including XWANDA, Fuxing Co., and Guolian Minsheng, showing divergence rates of 5.64%, 3.70%, and 3.15% respectively [1] - Stocks with smaller divergence rates that just crossed the annual line include Wangfujing, Tonghe Pharmaceutical, and Sanxia Water Conservancy [1] Top Stocks by Divergence Rate - The top three stocks with the highest divergence rates are: - XWANDA (9.13% increase, 5.12% turnover rate, latest price 21.16 yuan, divergence rate 5.64%) [1] - Fuxing Co. (4.31% increase, 5.52% turnover rate, latest price 2.42 yuan, divergence rate 3.70%) [1] - Guolian Minsheng (6.66% increase, 3.68% turnover rate, latest price 11.21 yuan, divergence rate 3.15%) [1] Additional Stocks with Notable Performance - Other stocks with significant performance include: - Tuoer Si (3.05% increase, 3.72% turnover rate, latest price 18.56 yuan, divergence rate 2.82%) [1] - ST Huaming (4.83% increase, 10.28% turnover rate, latest price 10.21 yuan, divergence rate 2.73%) [1] - Green Beauty (3.01% increase, 2.26% turnover rate, latest price 6.50 yuan, divergence rate 2.32%) [1]

Group 1 - Huawei Pangu concept declined by 2.40%, ranking among the top declines in the concept sector as of June 10 [1] - The concept sector saw a net outflow of 8.09 billion yuan, with 24 stocks experiencing net outflows, and 9 stocks seeing outflows exceeding 30 million yuan [2] - The stock with the highest net outflow was Tuowei Information, with a net outflow of 1.76 billion yuan [2] Group 2 - The top gainers in the Huawei Pangu concept included Meiansen and Jiecheng Shares, which rose by 1.72% and 0.19% respectively [1][3] - The concept sector's performance was contrasted with other sectors, such as the Transgenic sector which gained 3.15% [2] - The stocks with the highest net inflow included Meiansen, Huakai Yibai, and Fanwei Network, with inflows of 285.5 million yuan, 6.71 million yuan, and 4.69 million yuan respectively [2][3]

Core Insights - The natural language processing (NLP) industry in China is projected to reach a market size of approximately 12.6 billion yuan in 2024, reflecting a year-on-year growth of 14.55% [1][15] - The cost of model training has significantly decreased due to the "East Data West Computing" initiative, which provides low-cost computing power, and the adoption of few-shot learning frameworks has reduced the demand for training data by 90% [1][15] - Major companies in the NLP sector include Baidu, iFlytek, and Alibaba, each leveraging their technological strengths to capture market share in various applications [2][17][21] Industry Overview - NLP is a crucial branch of computer science and artificial intelligence, aimed at enabling computers to understand, interpret, and generate human language [1][8] - The technology types in NLP are primarily categorized into rule-based methods, statistical methods, and deep learning methods [1][8] Industry Development History - The development of NLP in China has gone through four main stages: the initial phase (1950s-60s) focused on machine translation, the rule-dominated phase (1970s-80s) involved complex rule systems, the statistical learning phase (1990s-2012) integrated statistical models with machine learning, and the deep learning phase (2013-present) is characterized by the dominance of deep learning models and pre-trained language models [4][5][6] Industry Value Chain - The upstream of the NLP industry chain includes hardware devices, data services, open-source models, and cloud services, while the midstream focuses on NLP technology research and development, and the downstream encompasses applications in finance, healthcare, education, and smart manufacturing [1][8] Market Size - The NLP industry in China is experiencing significant growth, with a projected market size of 12.6 billion yuan in 2024, driven by advancements in pre-trained language models and reduced training costs [1][15] Key Companies' Performance - Baidu leads the NLP industry with a strong technological foundation and extensive commercialization, maintaining the largest market share [17][21] - iFlytek excels in voice recognition and machine translation, particularly in the education and healthcare sectors [17][20] - Alibaba has made breakthroughs in machine reading comprehension and natural language understanding, integrating its technology into various business scenarios [17][20] Industry Development Trends - The NLP industry is witnessing a trend towards the integration of large models and multimodal capabilities, enhancing performance and user interaction [24] - There is a growing focus on vertical applications in sectors like healthcare and finance, as well as the integration of NLP with smart hardware [26] - Data security and ethical standards are becoming increasingly important, driving sustainable development in the NLP sector [27]

Core Insights - The article discusses the core technologies of multimodal large models, focusing on representation learning, translation, alignment, fusion, and collaborative learning [1][2][7][11][14]. Representation Learning - Representation learning is fundamental for multimodal tasks, addressing challenges such as combining heterogeneous data and handling varying noise levels across different modalities [1]. - Prior to the advent of Transformers, different modalities required distinct representation learning models, such as CNNs for computer vision (CV) and LSTMs for natural language processing (NLP) [1]. - The emergence of Transformers has enabled the unification of multiple modalities and cross-modal tasks, leading to a surge in multimodal pre-training models post-2019 [1]. Translation - Cross-modal translation aims to map source modalities to target modalities, such as generating descriptive sentences from images or vice versa [2]. - The use of syntactic templates allows for structured predictions, where specific words are filled in based on detected attributes [2]. - Encoder-decoder architectures are employed to encode source modality data into latent features, which are then decoded to generate the target modality [2]. Alignment - Alignment is crucial in multimodal learning, focusing on establishing correspondences between different data modalities to enhance understanding of complex scenarios [7]. - Explicit alignment involves categorizing instances with multiple components and measuring similarity, utilizing both unsupervised and supervised methods [7][8]. - Implicit alignment leverages latent representations for tasks without strict alignment, improving performance in applications like visual question answering (VQA) and machine translation [8]. Fusion - Fusion combines multimodal data or features for unified analysis and decision-making, enhancing task performance by integrating information from various modalities [11]. - Early fusion merges features at the feature level, while late fusion combines outputs at the decision level, with hybrid fusion incorporating both approaches [11][12]. - The choice of fusion method depends on the task and data, with neural networks becoming a popular approach for multimodal fusion [12]. Collaborative Learning - Collaborative learning utilizes data from one modality to enhance the model of another modality, categorized into parallel, non-parallel, and hybrid methods [14][15]. - Parallel learning requires direct associations between observations from different modalities, while non-parallel learning relies on overlapping categories [15]. - Hybrid methods connect modalities through shared datasets, allowing one modality to influence the training of another, applicable across various tasks [15].

Core Insights - The multi-modal large model market in China is projected to reach 15.63 billion yuan in 2024, an increase of 6.54 billion yuan from 2023, and is expected to grow to 23.48 billion yuan in 2025, indicating strong market demand and government support [1][6][19] Multi-Modal Large Model Industry Definition and Classification - Multi-modal large models are AI systems capable of processing and understanding various data forms, including text, images, audio, and video, using deep learning technologies like the Transformer architecture [2][4] Industry Development History - The multi-modal large model industry has evolved through several stages: task-oriented phase, visual-language pre-training phase, and the current multi-modal large model phase, focusing on enhancing cross-modal understanding and generation capabilities [4] Current Industry Status - The multi-modal large model industry has gained significant attention due to its data processing capabilities and diverse applications, with a market size projected to grow substantially in the coming years [6][19] Application Scenarios - The largest application share of multi-modal large models is in the digital human sector at 24%, followed by gaming and advertising at 13% each, and smart marketing and social media at 10% each [8] Industry Value Chain - The industry value chain consists of upstream components like AI chips and hardware, midstream multi-modal large models, and downstream applications across various sectors including education, gaming, and public services [10][12] Competitive Landscape - Major players in the multi-modal large model space include institutions and companies like the Chinese Academy of Sciences, Huawei, Baidu, Tencent, and Alibaba, with various models being developed to optimize training costs and enhance capabilities [16][17] Future Development Trends - The multi-modal large model industry is expected to become more intelligent and humanized, providing richer and more personalized user experiences, with applications expanding across various fields such as finance, education, and content creation [19]

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].