算法

Search documents
算法“开箱”让“杀熟”曝光
Guang Zhou Ri Bao· 2025-04-29 21:30
Group 1 - The core issue of "big data killing familiarity" involves platforms using personal data to implement differential pricing, leading to consumer dissatisfaction [1] - Consumers report being charged more for the same products in live-streaming sessions, highlighting a lack of transparency in pricing [1] - The collection of personal information and the subsequent analysis of consumer habits are the preliminary steps that enable differential pricing strategies [1] Group 2 - Algorithms should not operate outside of governance; their design and application must align with human-centered principles to protect consumer rights [2] - There is a call for platforms to adopt an open approach, enhancing algorithm transparency to combat "big data killing familiarity" [2] - Recommendations include returning control of information recommendations to users and establishing industry standards to improve algorithm transparency [2]
从“黑盒”到“手机爹”,一段人类驯服算法简史
远川研究所· 2025-04-29 12:42
技术进步日新月异,但人类的整活能力总在领先一步。 2025年至今简中互联网最火的流行语,一是"国运论",这篇透过deepseek看未来的雄文,后来被证实是AI生 成的内容; 二是"手机爹又把我干哪来了",表达一种下滑刷到陌生内容的诧异感,延伸出来还有大量表情包,甚至短视 频滤镜。 每一条火爆全网的内容下,都有一群仿佛走错房间的网友,拘谨又嚣张地带着"手机爹"的迷茫表情包,光速 加入互动:上一秒还在看中科院院士讲解黑洞合并,下一秒就被甲亢哥带着欣赏广场舞;刚看了两集《450 分钟深度解读红楼梦》,转头就迷失在修驴蹄子的白噪音中。 硅基大脑能俯仰古今畅聊国运,但将"手机"和"爹"排列组合到一起,是只有碳基大脑能整出的绝活。这背 后,是技术与人的关系正在悄然改变。 当快速更迭的推荐算法,向我们展现出一个更多面的世界,大部分人都愿意放下刻板成见,借助代码踏进 未曾想象过的广阔天地。 贝索斯在后来给股东的信中写道:" 我们的愿景是,让世界上每一本书,无论语言如何,都能在60秒内获取 [1]。 " 这位前世界首富的出发点显然没那么简单。对电商平台来说,图书是不可多得的标品品类,电子书更完美 解决了唯一不足的库存负担。 但 ...
兆威机电vs 雷赛智能:灵巧手之争|智能涌现
2 1 Shi Ji Jing Ji Bao Dao· 2025-04-29 04:26
Core Viewpoint - The competition between two leading companies in the industrial automation sector, Zhaowei Electromechanical and Raise Intelligent, is intensifying in the humanoid robot market, particularly focusing on the "dexterous hand" technology, which is considered the most valuable part of the upper body of humanoid robots [1][2]. Financial Performance - In 2024, Zhaowei Electromechanical achieved a revenue of 1.525 billion yuan, representing a year-on-year growth of 26.42%, with a net profit of 225 million yuan, up 25.11%. The overall gross margin was 31.43% [1]. - Raise Intelligent reported a revenue of 1.584 billion yuan, with a year-on-year increase of 11.93%, and a net profit of 200 million yuan, which grew by 44.67%. The overall gross margin was 38.27% [1]. R&D Investment - Both companies maintained high levels of R&D investment in 2024, with Zhaowei Electromechanical investing 155 million yuan, accounting for 10.18% of its total revenue, and holding 87 invention patents [1]. - Raise Intelligent invested 195 million yuan in R&D, representing 12.30% of its total revenue, with over 540 patents in its portfolio [1]. Product Differentiation - Zhaowei Electromechanical focuses on high precision and bionic features in its dexterous hand products, utilizing a micro harmonic reducer for 0.05mm positioning accuracy and 17-20 degrees of freedom [2]. - Raise Intelligent emphasizes high load capacity and industrial adaptability, with its DH116 dexterous hand capable of handling loads up to 40kg, targeting logistics and automotive assembly applications [2]. Market Positioning - Zhaowei Electromechanical positions itself as a high-end industrial dexterous hand solution provider, focusing on high freedom, high precision, and human-like products [2]. - Raise Intelligent positions itself as a supplier of core components and modular solutions for robots, targeting the mid-to-low-end market [2]. Supply Chain Strategy - Zhaowei Electromechanical adopts a "vertical integration" strategy, emphasizing the combination of AI algorithms with hand movement control [3]. - Raise Intelligent focuses on "core component breakthroughs," expanding its distribution network and increasing the proportion of its distribution partners [3]. Subsidiary Development - In 2024, both companies established subsidiaries to enter the robotics sector, with Raise Intelligent creating two subsidiaries focused on high-density motors and dexterous hand solutions [3]. - Zhaowei Electromechanical established a subsidiary in the U.S. and previously set up one in Germany [3]. Globalization Efforts - Zhaowei Electromechanical has a faster global expansion, with 14% of its revenue coming from international markets in 2024, growing at 35.15% [4]. - Raise Intelligent's international revenue accounted for only 4% of its total revenue in 2024 [4]. Competitive Outlook - Zhaowei Electromechanical is expected to leverage technological innovation and industrial-grade performance to gain an advantage in the high-end market [5]. - Raise Intelligent's modular solutions and cost control capabilities may provide a competitive edge in the mid-to-low-end market [5].
用户自选内容推荐强度 能破“信息茧房”?
Nan Fang Du Shi Bao· 2025-04-29 04:20
Core Viewpoint - Recent updates in apps like Xiaohongshu and Weibo allow users to visualize and adjust their content preferences, addressing concerns about algorithm transparency and user control over recommendations [1][4]. Group 1: User Control Features - Xiaohongshu introduced a "content preference adjustment" feature that displays a pie chart of users' browsing preferences over the past week, allowing them to adjust recommendation intensity across categories such as fashion, food, and travel [2][3]. - Users can choose from three levels of recommendation intensity: "reduce recommendation," "default intensity," and "increase intensity," and can also delete or add content tags [2][3]. - Similar features have been implemented in Weibo and Bilibili, where users can manage their content preferences through a pie chart and adjust recommendation settings [2][3]. Group 2: Algorithm Governance and Concerns - There are growing concerns about algorithmic issues such as "killing familiarity" and "information cocooning," prompting users to call for more control over algorithmic recommendations [4]. - The implementation of the "Internet Information Service Algorithm Recommendation Management Regulations" in 2022 mandates that platforms provide options for users to opt-out of personalized recommendations and manage their user tags [4]. - In response to these concerns, major platforms have introduced options for users to toggle personalized recommendations, aligning with regulatory requirements [5]. Group 3: Regulatory Actions - A joint initiative by four ministries launched in November aims to address issues like "information cocooning" and algorithmic manipulation, with a focus on user rights and algorithm transparency [5]. - The initiative requires platforms to conduct self-assessments and report on their compliance with the regulations, which has led to various governance measures being announced by platforms like Douyin, Weibo, and Xiaohongshu [5].
4.29犀牛财经早报:ChatGPT杀入电商赛道 幸福航空五一假期航班全部取消
Xi Niu Cai Jing· 2025-04-29 02:26
Group 1 - The public offering of index-enhanced funds is experiencing explosive growth, with new issuances this year reaching nearly seven times that of the same period last year, despite recent stagnation in fund size growth [1] - The stagnation is attributed to high investor cognitive barriers and the instability of excess returns, prompting fund companies to optimize strategies and innovate services to explore new opportunities in index investment [1] Group 2 - Alibaba has open-sourced its new Qwen3 model, which has one-third the parameters of DeepSeek-R1, significantly reducing costs while outperforming leading models [1] - The Qwen3 model integrates "fast thinking" and "slow thinking" within a single framework, utilizing a mixture of experts architecture, with a total parameter count of 235 billion and activation requiring only 22 billion [1] Group 3 - OpenAI is entering the e-commerce space by allowing users to purchase products through ChatGPT, redirecting them to merchant websites for checkout [2] - The product recommendations are based on user preferences and reviews from across the web, with OpenAI not charging affiliate marketing fees for purchases made through ChatGPT [2] Group 4 - The oil service industry is facing challenges due to falling international oil prices, but companies like CNOOC Services and Jereh are reporting stable performance in their Q1 2025 results [3] - Oil service companies are diversifying their operations to enhance competitiveness in response to potential risks from oil price fluctuations [3] Group 5 - Several A-share companies are changing their stock names to reflect strategic transformations and enhance brand image, with over 20 companies reported to have announced name changes this year [8] - The name changes are seen as a way to attract investor attention and signal new directions for the companies [8] Group 6 - Foton Motor announced plans to subscribe for up to 2.5 billion yuan worth of shares in Beiqi Blue Valley, which is raising up to 6 billion yuan in total [9] - This transaction is classified as a related party transaction and does not constitute a major asset restructuring [9] Group 7 - Greenland Holdings reported a net loss of 247 million yuan in Q1 2025, with revenues declining by 30.97% year-on-year [10] - Gree Real Estate also reported a loss of 90.96 million yuan in Q1 2025, with revenues down 58.54% compared to the previous year [11]
阿里Qwen3性能超越DeepSeek-R1;美媒曝马斯克孩子数量远超14个;ChatGPT推出购物功能
Guan Cha Zhe Wang· 2025-04-29 01:10
Group 1: Stock Market Performance - The three major U.S. stock indices closed mixed, with the Dow Jones up by 0.28% and the S&P 500 up by 0.06%, while the Nasdaq fell by 0.1% [1] - Major tech stocks showed varied performance, with Intel rising over 2%, while Nvidia dropped over 2% [1] Group 2: AI and Technology Developments - Alibaba's Qwen3 has been released as an open-source model, surpassing competitors with a total of 235 billion parameters [2] - Apple CEO Tim Cook has reorganized the company's robotics team, indicating dissatisfaction with the progress in AI and machine learning [6][5] - OpenAI is enhancing its ChatGPT tool to facilitate online shopping, allowing users to purchase products directly through the platform [7] Group 3: Investment and Financial Activities - Amazon launched its first batch of satellites for the "Project Kuiper" internet initiative, aiming to deploy over 3,200 satellites for global internet coverage [7] - Alphabet plans to issue approximately $4 billion in high-grade corporate bonds, with the longest maturity potentially yielding 1% to 1.05% above U.S. Treasury rates [7] - Over 700 billion yuan has been allocated by local government funds towards humanoid robotics and related industries [8] Group 4: Company Listings and IPOs - Seres has applied for a mainboard listing in Hong Kong, with projected revenue of 145.1 billion yuan in 2024, marking a 305.5% year-on-year increase [9] - Stone Technology is considering an IPO in Hong Kong to raise up to $500 million, although plans are still in the early stages [10]
四大证券报精华摘要:4月29日
Zhong Guo Jin Rong Xin Xi Wang· 2025-04-28 23:56
Group 1 - The National Market Regulation Administration has approved the release of two revised national standards for tourism services, which will enhance the tourism service standard system and stimulate the tourism consumption market [1] - Over 20 A-share companies have announced changes to their stock names, reflecting strategic transformations and aiming to attract investor attention [1] Group 2 - Multiple fund companies have received notifications about a 20% reduction in index usage fees from the China Securities Index Company, effective from April 1 [2] - The quarterly minimum fee for index products has been lowered to 20,000 yuan, with no minimum for certain non-stock index products [2] Group 3 - The scale of ETF funds has surpassed 4 trillion yuan, intensifying competition in the pure index sector, while the growth of index-enhanced funds has recently stagnated [3] - Industry insiders attribute the stagnation to high investor cognitive barriers and the instability of excess returns [3] - Oil service companies in A-shares report stable performance despite recent declines in international oil prices, with companies like CNOOC and Jereh showing solid results [3] Group 4 - The pace of new IPO approvals in A-shares has accelerated, indicating a gradual recovery in the IPO market, with expectations for slight growth in 2025 [4] - Deloitte China predicts that the new stock issuance in 2025 will focus on quality and support for technological innovation [4] Group 5 - The securities industry is expected to see a reversal of difficulties in 2024, with revenue and net profit growth, although investment banking services are still struggling [5] - Only 6 out of 38 listed brokerages reported positive growth in investment banking fees, with an average decline of 27% for the others [5] Group 6 - The "Guzi Economy" concept stock, Pop Mart, has seen significant stock price increases, becoming the first tenfold stock in this sector [6] - The transparency of algorithms used by major platforms is increasing, with implications for digital governance and business models [6] Group 7 - The National Development and Reform Commission has outlined measures to stabilize employment and promote high-quality economic development, focusing on effective investment and consumer support [7] - 15 out of 17 listed banks reported positive year-on-year growth in investment income for Q1 2025, with some banks showing over 100% growth [7] Group 8 - Beijing SKP is among the first stores to implement the "buy and return immediately" tax refund service, which is being expanded nationwide to enhance the consumer experience for international travelers [8]
十亿分之一克水体污染物,1分钟“揪出”!
Ke Ji Ri Bao· 2025-04-28 22:46
Core Insights - The China Agricultural Academy of Agricultural Quality Standards and Testing Technology Research Institute has developed the world's first intelligent monitoring system for agricultural non-point source pollution, which enables real-time monitoring of high-risk antibiotics and pesticide residues in flowing water bodies [1][2] Group 1: Technological Breakthroughs - The research team led by Cheng Jie has achieved breakthroughs in materials, technology, and equipment to address the challenges of monitoring non-point source pollution, which is characterized by its dispersion, concealment, and complexity [2] - A new type of nano-composite material with molecular recognition and signal enhancement capabilities has been developed, improving stability by over five times compared to traditional adsorbents and significantly enhancing sensing performance [3] - The system can detect three categories of nine pesticides and three categories of ten antibiotics with a sensitivity of nanograms per milliliter (ng/mL), allowing for the detection of pollutants at concentrations as low as one billionth of a gram in one milliliter of water [3] Group 2: Operational Efficiency - The monitoring system can operate continuously for 180 days without human intervention, reducing operational costs by 70% and significantly improving monitoring efficiency [3][4] - In a demonstration project in the Haihe River basin, labor costs were reduced by approximately 60%, while monitoring frequency increased threefold [4] Group 3: Application and Impact - The system has been successfully applied in the Taihu and Haihe River basins, achieving an accuracy rate of over 90% in pollution warnings through real-time data uploads to a cloud platform and GIS technology [5] - The system allows for precise pollution source tracing, aiding management departments in implementing targeted remediation measures [6] - Future enhancements will include the detection of new pollutants such as perfluorinated compounds, endocrine disruptors, and microplastics, expanding the monitoring scope [6]
平台算法从“黑箱”走向透明
Zheng Quan Shi Bao· 2025-04-28 22:28
算法破"茧" 数字化和人工智能时代,算法渗透无处不在。它既可以引向高效率和美好愿景,也可能暗藏"毒流量"和 被制定者利用。 4月中旬,抖音算法首度公开,在业内掀起不小的波澜。长期跟踪TMT领域的投资经理杨勇对记者分 析,此前平台曾出现吴柳芳事件等多例争议事件,抖音方面虽然有所回应,但始终未正面披露其背后的 算法机制。这次公开,在所有互联网平台里都堪称"大尺度"。 从去年底至今,抖音、小红书、美团等越来越多的互联网平台推出了新一轮算法公开化举措,把驯服算 法的主动权交到用户手中。 近期,抖音首次公开推荐算法原理。 时报财经图库/供图 去年底至今,小红书等平台推出了新一轮算法公开化举措。 时报财经图库/供图 证券时报记者 王小伟 从骑手运力的分配到酒店住宿的推荐,从网约车辆的调度到电商平台的供应链管理,算法已成为所有大 厂平台的底层工具,也是中国互联网世界的"隐形舵手"。 今年以来,算法一改"神秘科技"属性,透明化进程提速。根据证券时报记者梳理,除少数平台外,抖 音、美团、小红书等平台都已将算法"黑箱"越开越大。在此过程中,一些"反常识"认知的计算方式也陆 续出现。 主流判断认为,平台算法治理是互联网企业2025 ...
机器学习算法可解析纳米晶体结构
news flash· 2025-04-28 22:19
Core Insights - A machine learning algorithm developed by a team at Columbia University can infer the atomic structure of materials by observing patterns produced by nanocrystals, addressing a century-old challenge in materials science [1] Group 1: Implications for Drug Development - The breakthrough in nanocrystal structure analysis is expected to accelerate new drug development processes [1] Group 2: Clean Energy Materials - The new algorithm may also facilitate the development of clean energy materials, potentially impacting the renewable energy sector positively [1] Group 3: Cultural Heritage Research - This advancement could enhance research in cultural heritage, providing new methods for analyzing and preserving historical materials [1]