瑞银经济学家Amanda Wilcox在上周五的一份报告中写道："库格勒理事席位的提前空缺，将使下任主 席的遴选工作进入加速阶段。"美联储此前宣布，库格勒将于8月8日正式离任。这位2023年9月就任、 原定任期至2026年1月的理事，上周因个人原因缺席了联邦公开市场委员会(FOMC)会议。 库格勒在任职未满两年之际提前离任，正值白宫酝酿鲍威尔最终继任方案的关键时刻。 点击蓝字，关注我们 据瑞银经济学家分析，美联储理事库格勒上周五的意外辞职，使得主席鲍威尔继任者的遴选工作被迫提 前启动 ，这一职位空缺可能导致美联储领导层重组远早于预期。 而关于鲍威尔的潜在继任者，媒体和政策界已有诸多猜测。Wilcox指出："包括《华尔街日报》6月报 道在内的多家媒体已披露若干候选人名单，财政部长斯科特·贝森特、国家经济委员会主任凯文·哈西特 和前理事凯文·沃什均在列。我们认为本周两位支持降息25个基点的异议者之一沃勒理事也可能参与角 逐。" 尽管瑞银未对政府最终决策作出预测，但Wilcox强调："我们期待关注这一进程，特别是在今天下午的 辞职事件使得时间表明显提前之后。" 库格勒理事的提前离任使美联储理事会出现七个空缺席位，白 ...

Core Viewpoint - The collaboration between two Tsinghua University alumni has successfully enhanced the Gemini 2.5 Pro model to achieve a gold medal level in the International Mathematical Olympiad (IMO) through a self-iterative verification process and prompt optimization [1][4][10]. Group 1: Model Performance and Methodology - Gemini 2.5 Pro achieved a 31.55% accuracy rate in solving IMO problems, significantly outperforming other models like O3 and Grok 4 [9]. - The research team utilized a structured six-step self-verification process to improve the model's performance, which includes generating initial solutions, self-improvement, and validating solutions [16][18]. - The model was able to generate complete and mathematically rigorous solutions for 5 out of 6 IMO problems, demonstrating the effectiveness of the structured iterative process [24][23]. Group 2: Importance of Prompt Design - The use of specific prompt designs significantly improved the model's ability to solve complex mathematical problems, highlighting the importance of prompt engineering in AI model performance [12][14]. - The research indicated that detailed prompts could reduce the computational search space and enhance efficiency without granting the model new capabilities [23]. Group 3: Research Team Background - The authors, Huang Yichen and Yang Lin, are both Tsinghua University alumni with extensive academic backgrounds in physics and computer science, contributing to the credibility of the research [26][28][33]. - Yang Lin is currently an associate professor at UCLA, focusing on reinforcement learning and generative AI, while Huang Yichen has a strong background in quantum physics and machine learning [30][35]. Group 4: Future Directions and Insights - The research team plans to enhance the model's capabilities through additional training data and fine-tuning, indicating a commitment to ongoing improvement [42]. - Yang Lin expressed the potential for AI to play a more significant role in mathematical research, especially in addressing long-standing unresolved problems [44].

Group 1 - The core point of the article is that Apple reported a strong Q3 FY2025 financial performance, with total revenue reaching $98 billion, a 10% year-over-year increase, marking the largest quarterly revenue growth since Q1 FY2022 [1][3] - The Mac and iPad segments performed well, with Mac revenue increasing by 20% to $8.5 billion, driven by the popularity of the new MacBook Air with M4 chip, and iPad revenue rising by 18% to $7 billion, with a 30% surge in education market purchases [3] - The Greater China region, Apple's third-largest market, generated $17 billion in revenue, a 7% year-over-year increase, ending a streak of four consecutive quarters of decline [3] Group 2 - Apple CEO Tim Cook outlined the company's AI strategy, indicating a significant increase in capital expenditure and R&D investment in generative AI and machine learning, focusing on enhancing personalized experiences while protecting user privacy [3] - Cook emphasized that AI is deeply integrated into every new product, from Siri optimization to photo editing and health monitoring, redefining device interaction through edge AI [3] - Analysts noted Cook's openness to acquisitions in the AI space, suggesting that Apple may pursue technology-driven acquisitions to enhance its AI capabilities and catch up with competitors like Microsoft and Google [3][4] Group 3 - Some analysts raised concerns about the pace of Apple's AI strategy implementation, with Morgan Stanley pointing out the lack of details on self-developed large models and the reliance on third-party models for AI functionalities [4] - Goldman Sachs raised Apple's target price to $240, believing that the combination of edge AI and privacy protection will attract high-end users and further improve service business margins [4]

Financial Data and Key Metrics Changes - The company reported total revenue of $186 million, representing a 27% increase year over year, with surgical revenue growing by 29% to $168 million [9][10] - Adjusted EBITDA reached a record $23 million, accounting for 13% of revenue, marking an improvement of 880 basis points year over year [4][16] - Free cash flow was $5 million, indicating a positive cash generation trend [4][18] Business Line Data and Key Metrics Changes - Surgical revenue growth was driven by a procedural volume increase of 28%, with surgeon adoption growing by 21% and utilization increasing by 6% [10][11] - EOS revenue increased by 11% year over year, contributing $17 million to total revenue [13] - Same store sales in established territories grew by 29%, reflecting strong demand in existing markets [5][11] Market Data and Key Metrics Changes - The company has achieved a market share ranking of third in the U.S. spine market, indicating significant competitive positioning [7][38] - The company continues to grow at five to six times the overall market rate, showcasing its strong market presence [22] Company Strategy and Development Direction - The company is focused on creating clinical distinction and enhancing surgeon adoption through innovative product offerings and a robust sales force [24][30] - Investments in technology infrastructure are aimed at supporting long-term growth and profitability, with a focus on integrating various surgical tools into a cohesive ecosystem [30][34] - The company plans to launch a new robotic system in early 2026, which will be integrated into existing surgical workflows [65][66] Management's Comments on Operating Environment and Future Outlook - Management expressed confidence in the company's ability to maintain positive cash flow and profitability, with expectations for continued strong revenue growth [20][19] - The company anticipates a sequential step down in revenue from Q2 to Q3, typical for the seasonality of the business [98] - Management highlighted the importance of operational improvements and asset management in driving future profitability [19][20] Other Important Information - The company has raised its full-year revenue guidance by $8 million to $742 million, reflecting strong performance in the surgical business [20][63] - Non-GAAP gross margin was reported at 70%, with a slight decrease year over year due to product mix changes [14] Q&A Session Summary Question: What differentiates the upcoming robotic system? - The company emphasized the integration of navigation robotics into the surgical workflow, aiming for increased precision and efficiency in spine procedures [41][46] Question: How much of the same store growth is attributed to new reps and instrument availability? - Management noted that both new hires and increased instrument availability have contributed to growth, with a focus on surgeon adoption and utilization driving results [51][56] Question: What are the expectations for organic growth and geographic penetration? - The company expects over 20% organic growth, with a focus on expanding in under-indexed geographies and leveraging the EOS technology for predictive analytics [62][94] Question: What is the outlook for CapEx and free cash flow? - Management indicated a commitment to positive free cash flow while maintaining strategic CapEx investments to support growth initiatives [109]

Group 1 - Micron Technology's stock price decreased by 5.07% to $108.92 per share, with a trading volume of $1.213 billion and a total market capitalization of $121.895 billion [1] - For the fiscal year ending May 29, 2025, Micron Technology is projected to have total revenue of $26.063 billion, representing a year-over-year growth of 50.12%, and a net profit attributable to shareholders of $5.338 billion, showing a staggering increase of 4997.25% [1] - Micron Technology is a global leader in the semiconductor industry, offering a wide range of high-performance memory and storage technologies, including DRAM, NAND, NOR Flash, and 3D XPoint memory [1] Group 2 - The company has a 40-year history of technological leadership, with its memory and storage solutions driving disruptive trends in key market areas such as cloud data centers, networking, mobile, artificial intelligence, machine learning, and autonomous vehicles [1] - Micron's common stock (MU) is traded on the NASDAQ exchange [1] - The company is scheduled to disclose its fiscal year 2025 annual report on September 24, with the actual release date subject to company announcement [1]

智通财经APP获悉，随着全球范围内数千亿美元资金投入电网现代化改造与扩建工程，以满足数字化与 清洁能源革命带来的巨大电力需求，铜消费增速已远超行业预期。然而，智利、刚果(金)等主要产铜国 因新矿投资不足导致供应受限，这为铜价长期高位运行埋下伏笔。部分分析师预测，2030年前铜价将突 破每吨1.2万美元的历史峰值，较当前约9700美元/吨的价格上涨23%。 尽管终端用户正寻求替代方案，但铜因其卓越的导电性、耐用性和多功能性仍难以被取代。国际能源署 数据显示，全球电网投资额继2024年创下3900亿美元纪录后，今年预计将突破4000亿美元大关。 咨询公司Benchmark Mineral Intelligence(BMI)战略总监Michael Finch表示，"铜在电网基础设施中常被严 重低估。虽然各国都意识到扩建电网的必要性，却普遍低估了所需的铜材总量"，特别是美国、英国和 中国这三大关键市场的投资需求。 据该机构提供的最新预测，全球发电与输电网络升级带来的铜需求将从今年的1252万吨增至2030年的 1487万吨。 数据中心与电动汽车成需求新引擎 美国银行分析师Michael Widmer预计，到2030年 ...

Quantitative Factors and Construction Factor Name: GAN_GRU Factor - **Construction Idea**: The GAN_GRU factor is derived by processing volume-price time-series features using a Generative Adversarial Network (GAN) model, followed by encoding these time-series features with a Gated Recurrent Unit (GRU) model to generate a stock selection factor [4][13][41] - **Construction Process**: 1. **Input Features**: 18 volume-price features such as closing price, opening price, turnover, and turnover rate are used as input data. These features are sampled every 5 trading days over the past 400 days, resulting in a feature matrix of shape (40,18) [14][17][18] 2. **Data Preprocessing**: - Outlier removal and standardization are applied to each feature over the 40-day time series - Cross-sectional standardization is performed at the stock level [18] 3. **GAN Model**: - **Generator**: An LSTM-based generator is used to preserve the sequential nature of the input features. The generator takes random noise (e.g., Gaussian distribution) as input and generates data that mimics the real data distribution [23][33][37] - **Discriminator**: A CNN-based discriminator is employed to classify real and generated data. The discriminator uses convolutional layers to extract features from the 2D volume-price time-series "images" [33][35] - **Loss Functions**: - Generator Loss: $$ L_{G} = -\mathbb{E}_{z\sim P_{z}(z)}[\log(D(G(z)))] $$ where \( z \) represents random noise, \( G(z) \) is the generated data, and \( D(G(z)) \) is the discriminator's output probability for the generated data being real [24] - Discriminator Loss: $$ L_{D} = -\mathbb{E}_{x\sim P_{data}(x)}[\log D(x)] - \mathbb{E}_{z\sim P_{z}(z)}[\log(1-D(G(z)))] $$ where \( x \) is real data, \( D(x) \) is the discriminator's output probability for real data, and \( D(G(z)) \) is the discriminator's output probability for generated data [27] 4. **GRU Model**: - Two GRU layers (GRU(128,128)) are used to encode the time-series features, followed by an MLP (256,64,64) to predict future returns [22] 5. **Factor Output**: The predicted returns (\( pRet \)) from the GRU+MLP model are used as the stock selection factor. The factor is neutralized for industry and market capitalization effects and standardized [22] Factor Evaluation - The GAN_GRU factor effectively captures the sequential and cross-sectional characteristics of volume-price data, leveraging the strengths of GANs for feature generation and GRUs for time-series encoding [4][13][41] --- Factor Backtesting Results GAN_GRU Factor Performance Metrics - **IC Mean**: 11.43% (2019-2025), 10.97% (last year), 9.27% (latest month) [41][42] - **ICIR**: 0.89 [42] - **Turnover Rate**: 0.82 [42] - **Annualized Return**: 38.52% [42] - **Annualized Volatility**: 23.82% [42] - **IR**: 1.62 [42] - **Maximum Drawdown**: 27.29% [42] - **Annualized Excess Return**: 24.86% [41][42] GAN_GRU Factor Industry Performance - **Top 5 Industries by IC (Latest Month)**: - Home Appliances: 27.00% - Non-Bank Financials: 23.08% - Retail: 20.01% - Steel: 14.83% - Textiles & Apparel: 13.64% [41][42] - **Top 5 Industries by IC (Last Year)**: - Utilities: 14.43% - Retail: 13.33% - Non-Bank Financials: 13.28% - Steel: 13.23% - Telecommunications: 12.36% [41][42] GAN_GRU Factor Long Portfolio Performance - **Top 5 Industries by Excess Return (Latest Month)**: - Textiles & Apparel: 5.19% - Utilities: 3.62% - Automobiles: 3.29% - Non-Bank Financials: 2.56% - Pharmaceuticals: 1.47% [2][43] - **Top 5 Industries by Average Monthly Excess Return (Last Year)**: - Home Appliances: 5.44% - Building Materials: 4.70% - Textiles & Apparel: 4.19% - Agriculture: 4.09% - Utilities: 3.92% [2][43]

Group 1 - The core viewpoint of the article highlights significant developments in the automotive industry, including local production initiatives, new technology patents, and capacity expansion efforts by various companies [1][2][3] Group 2 - Haval M6 has officially launched full localization production in Russia, with a rental agreement signed for the PSMA Rus factory in Kaluga, aiming for full production by the end of 2024. The assembly process began in February, with modern upgrades to welding and painting facilities [1] - Lantu Motors announced that the new Lantu Zhiyin will feature 5C ultra-fast charging and an 800V platform, with over 20,000 pre-orders for the FREE+ model. To enhance production capacity and ensure timely delivery, Lantu will initiate a second factory with greater capacity than the first [1] - Changan Automobile has disclosed a patent for a driving behavior prediction method, utilizing advanced machine learning to accurately forecast driving behaviors, which will support advanced driver-assistance systems (ADAS) and intelligent driving technologies [1] - Xiaomi Auto has published a patent for a commuting reminder system that provides traffic status updates based on preset conditions, enhancing user experience by allowing for better planning of departure times [1]

- NVIDIA launched the OpenReasoning-Nemotron reasoning model series in July 2025, based on the Qwen2.5 architecture, distilled from the 671 billion-parameter DeepSeek R1 0528 model, and available in four parameter scales: 1.5B, 7B, 14B, and 32B. The model aims to support structured tasks such as mathematics, science, and code generation efficiently [12] - The core innovation of OpenReasoning-Nemotron lies in its data distillation strategy, leveraging the NeMo Skills framework to generate 5 million high-quality data trajectories covering mathematical proofs, scientific derivations, and programming solutions. The training process uses supervised fine-tuning (SFT) instead of reinforcement learning, ensuring logical consistency and precision in symbolic reasoning [12] - The model employs the GenSelect algorithm to implement a "heavy reasoning mode," which involves parallel generation of candidate solutions by multiple agents and selecting the optimal answer. For example, the GenSelect@64 on the 32B model improved HMMT math competition scores from 73.8 to 96.7 and enhanced LiveCodeBench scores from 70.2 to 75.3 in code generation tasks [13] - The OpenReasoning-Nemotron series achieved record-breaking results in benchmarks such as GPQA, MMLU-PRO, and AIME24. The 32B model scored 89.2 on AIME24, surpassing OpenAI's o3-high model, while the 7B model scored 78.2, representing a nearly 20% improvement over its predecessor. However, the 1.5B model showed performance degradation to 45.6 due to inconsistencies in handling 32K tokens [15] - The Qwen3-Coder model, developed by Alibaba Cloud's Tongyi Qianwen team, was officially open-sourced in July 2025. It features a 480 billion parameter scale with a native 256K context window and employs a sparse MoE design, activating only 35 billion parameters per inference. The model was trained on a 7.5 trillion token corpus, with 70% of the data being code, covering over 80 programming languages and 20 markup languages [19][20] - Qwen3-Coder achieved a HumanEval pass@1 accuracy of 93.7%, surpassing Claude 3.5's 92.4%. On the SWE-Bench Verified benchmark, it achieved a 31.4% task success rate, exceeding GPT-4's 30.9%. Key innovations include extending the native 256K context to 1M tokens using YaRN technology and integrating execution feedback mechanisms to validate and reward generated code [20] - The GitLab Duo platform, launched in public beta in July 2025, virtualizes traditional software development team roles into specialized AI agent clusters. These agents handle tasks such as requirement planning, code writing, security analysis, testing, and operations management, forming a dynamic collaboration network. The platform automates workflows through the "Flows" feature, enabling developers to input functional descriptions and have agents complete tasks like requirement decomposition, code generation, and testing [33][36] - GitLab Duo integrates with mainstream development environments like VS Code and JetBrains IDEs and plans to introduce a "knowledge graph" feature to enhance agents' understanding of code context. The platform also emphasizes security, employing end-to-end encryption and sandbox environments for code validation [36][37]

Core Viewpoint - A new wearable all-in-one obstructive sleep apnea management system has been developed, integrating flexible piezoelectric monitoring and soft magnetoelastic stimulation, addressing the limitations of traditional polysomnography (PSG) in cardiovascular parameter monitoring [4][11]. Group 1: Research Development - The research was a collaboration between Southwest Jiaotong University, City University of Hong Kong, and West China Hospital of Sichuan University [2]. - The developed system features a customized piezoelectric composite sensor for continuous physiological signal monitoring and a soft magnetoelastic actuator for non-invasive mechanical stimulation [7][11]. - The system utilizes a machine learning algorithm to achieve a 92.7% accuracy rate in real-time detection of sleep apnea events [8]. Group 2: Clinical Validation - Rigorous laboratory and clinical studies demonstrated that the developed apnea management system (AMS) is comparable to the clinical gold standard, PSG, in identifying apnea events [9]. - Parallel comparison signals from AMS and PSG confirmed the effectiveness of feedback stimulation [11]. Group 3: System Features and Benefits - The AMS integrates continuous physiological monitoring and non-invasive mechanical stimulation, providing a closed-loop system for sleep apnea management [11]. - This system not only addresses the limitations of traditional PSG in tracking cardiovascular responses but also offers a scalable and user-friendly platform for personalized sleep health care at home [11][12].