Overview - The article discusses the transformative impact of artificial intelligence (AI) and machine learning (ML) on marketing and supply chain management, emphasizing the need for businesses to adapt to these technologies for improved decision-making and operational efficiency [1][6]. AI Terminology Overview - AI encompasses a broad field focused on creating machines capable of tasks requiring human-like intelligence, while ML is a subset of AI that enables computers to learn from data without explicit programming [2][4]. Importance of AI - AI is being rapidly adopted across industries as it directly correlates with business efficiency, profitability, and competitiveness, moving beyond experimental phases to practical applications in daily operations [6][9]. Applications of AI in Marketing - AI is utilized in marketing through personalized recommendations, customer service chatbots, and predictive analytics, enhancing customer engagement and operational effectiveness [10][12]. Marketing's Impact on Supply Chain - Marketing activities can trigger demand shocks, necessitating a responsive supply chain to avoid stockouts and missed revenue opportunities, highlighting the interconnectedness of marketing and supply chain functions [13][15]. Challenges in Modern Supply Chains - Modern supply chains face challenges such as complexity, uncertainty, speed expectations, and sustainability, driving the need for AI to enhance demand forecasting and proactive measures [19][20]. AI in Demand Forecasting and Planning - AI enhances demand forecasting and planning by integrating time series analysis with machine learning, allowing for more accurate predictions and operational actions [20][22]. AI in Inventory Optimization - AI aids in inventory management by determining optimal stock levels based on real-time data and demand forecasts, balancing availability and cost [24][26]. AI in Logistics and Transportation - AI transforms logistics by optimizing delivery routes, predicting arrival times, and enabling predictive maintenance, thus improving efficiency and reliability [27][29]. AI in Supplier and Risk Management - AI strengthens supplier and risk management through continuous performance analysis and real-time monitoring of external events, allowing for proactive risk mitigation [33][34]. AI in Warehousing and Automation - AI automates and optimizes warehousing processes, improving accuracy and efficiency in inventory handling and order fulfillment [37][38]. AI in Sustainability and ESG - AI supports sustainability efforts by optimizing processes to reduce waste and emissions, facilitating the transition to circular supply chains [38][40]. Unified Perspective on Marketing and Supply Chain - Understanding AI's value requires viewing marketing and supply chain as interconnected systems, where AI synchronizes demand creation and fulfillment [61][63]. Emerging Trends in AI-Driven Supply Chains - New trends in AI include digital twins for simulation, proactive AI agents for planning, and visual models for real-time monitoring, indicating a shift towards more autonomous and intelligent supply chain operations [66][67].

Core Points - The ICML 2026 conference will take place from July 7 to July 12, 2026, in Seoul, South Korea, with a double-blind review process for all submitted papers [4] - Authors of accepted papers can choose whether to attend the conference in person or only have their papers included in the proceedings [7] - The original submission versions of accepted papers will be made publicly available, and authors of rejected papers can also choose to make their original submissions public [10] Submission Requirements - Papers must be submitted as a single file, with a maximum of 8 pages for the main text, while references, impact statements, and appendices have no page limit [5] - There will be no separate submission deadline for supplementary materials, and authors can add one extra page to the final version of accepted papers [6] - Papers that do not comply with the submission requirements will be rejected without review [11] Important Dates - The submission website will open on January 8, 2026, with the abstract submission deadline on January 23, 2026, and the full paper submission deadline on January 28, 2026 [14][15] Review Process - Authors are required to participate in the review process, with specific mutual review requirements for both papers and authors [17] - The double-blind review policy prohibits simultaneous submissions to multiple conferences or journals [18] - All submissions must be anonymized and should not contain any information that could reveal the authors' identities [21] Ethical Guidelines - Each paper must include a potential societal impact statement, which should be placed at the end of the paper and will not count towards the page limit [23] - Authors must submit a plain language summary to communicate the significance of their research to the public [24] - Violations of the review process or ethical guidelines may result in sanctions or rejection of the submission [22][23]

Group 1 - The core of the article revolves around the significant partnership between Microsoft and OpenAI, which aims to reshape the future of AI through a newly established agreement [3][8][60] - The partnership began in 2019 when Microsoft invested $1 billion in OpenAI, providing essential funding and cloud computing resources to support AI model training [5][7] - The recent agreement marks a new phase in their relationship, with OpenAI restructuring to create a Public Benefit Corporation (PBC) under its non-profit foundation, allowing for both profit and public good [8][10] Group 2 - OpenAI's foundation now holds shares valued at $130 billion, making it one of the largest charitable foundations globally, with plans to invest $25 billion in healthcare and AI safety [10][12] - Microsoft holds approximately 32.5% of OpenAI's shares, valued at around $135 billion, binding the fates of both companies together [15][16] - The partnership has evolved into one of the most successful collaborations in the industry, with both leaders expressing optimism about the future value of OpenAI [17][18] Group 3 - The new agreement includes exclusive deployment of OpenAI's advanced AI models on Microsoft's Azure cloud platform for the next seven years, making Azure a central hub for AI development [19][20] - Microsoft reported a 27% year-over-year revenue increase in its intelligent cloud segment, driven by Azure's growth, particularly in AI-related contracts [20][23] - OpenAI has committed to a $250 billion pre-purchase contract for Azure resources, ensuring ample computing power for its AI model training [20][22] Group 4 - Both companies face challenges related to computing power shortages, which have limited OpenAI's ability to onboard new users and expand its models [24][26] - Microsoft has significantly increased its capital expenditures to build data centers and acquire AI chips, yet still struggles to meet the soaring demand for computing resources [26][27] - The leaders predict that AI computing power will remain tight for the next few years, despite potential future advancements in technology [28][29] Group 5 - The partnership is also driving a transformation in software paradigms, with AI changing how users interact with applications, moving from traditional interfaces to conversational agents [33][34] - Microsoft is integrating AI capabilities into its Office products, enhancing their value and user engagement, while also exploring new business models for AI assistants [36][39] - The collaboration is expected to boost productivity and economic growth, with predictions of a potential return to 4% annual growth in the U.S. economy due to AI advancements [52][53] Group 6 - The partnership between Microsoft and OpenAI is not just about profit but also focuses on ensuring that AI benefits humanity as a whole [64][65] - Both companies are actively involved in shaping regulations and standards for AI to promote safe and responsible development [62][64] - The collaboration exemplifies a blend of idealism and pragmatism, aiming to harness technological innovation for the greater good [64]

Core Viewpoint - The article discusses the evolution of autonomous driving technology, highlighting the transition from rule-based systems to end-to-end models represented by companies like Ideal and XPeng, and currently to the world model phase represented by NIO, emphasizing the continuous presence of deep learning throughout these changes [1]. Group 1: Course Introduction - The course covers the development from modular production algorithms to end-to-end systems and now to VLA, focusing on core algorithms such as BEV perception, visual language models (VLM), diffusion models, reinforcement learning, and world models [5]. - Participants will gain a comprehensive understanding of the end-to-end technology framework and key technologies, enabling them to reproduce mainstream algorithm frameworks like diffusion models and VLA [5]. - Feedback indicates that students completing the course can achieve approximately one year of experience as end-to-end autonomous driving algorithm engineers, benefiting from the training for internships and job recruitment [5]. Group 2: Instructor Profile - The main instructor, Jason, holds a C9 undergraduate degree and a PhD from a QS top 50 university, with multiple published papers in CCF-A and CCF-B journals [6]. - He is currently an algorithm expert at a leading domestic manufacturer, engaged in the research and production of cutting-edge algorithms, with extensive experience in the development and delivery of autonomous driving perception and end-to-end algorithms [6]. Group 3: Research Guidance - The program aims to enhance practical skills and knowledge in cutting-edge topics, with a focus on helping students publish high-level papers to improve their academic prospects [8]. - The community includes over 300 instructors specializing in autonomous driving and embodied intelligence, with a high manuscript acceptance rate of 96% over the past three years [8]. Group 4: Research Process - The guidance process includes selecting research topics based on student interests, explaining key concepts, and providing essential foundational knowledge and recommended learning materials [11]. - Students will learn how to critically read literature, conduct research, and write various sections of a paper, including methods and experimental results, with continuous feedback and support throughout the process [11].

Core Insights - The AI industry is transitioning from a phase of rapid investment and growth to a more competitive and cost-conscious environment, as evidenced by layoffs and restructuring among major players like Meta, OpenAI, and AWS [1][2]. Group 1: Historical Context of AI Development - Google was founded with AI as a core principle, influenced by co-founder Larry Page's background in machine learning [5][9]. - The term "Artificial Intelligence" was first coined in 1956, but the field faced significant setbacks due to limitations in computing power and data, leading to two major "AI winters" [8]. - Larry Page's vision for Google included the belief that AI would be the ultimate version of their search engine, aiming to understand everything on the web [9][10]. Group 2: Key Innovations and Breakthroughs - Google's early AI efforts included the development of the PHIL language model, which significantly improved search functionalities and contributed to the company's revenue through AdSense [14][15][16]. - The introduction of neural networks and deep learning at Google was catalyzed by the arrival of key figures like Geoff Hinton, who advocated for the potential of deep learning [19][21]. - The "cat paper," which demonstrated a deep learning model's ability to recognize images without supervision, marked a significant milestone for Google Brain and had profound implications for YouTube's content understanding [30][34]. Group 3: Competitive Landscape and Strategic Moves - The success of AlexNet in 2012 revolutionized deep learning and established GPU as the core hardware for AI, leading to a surge in interest and investment in AI talent [35][39]. - Google acquired DNN Research, further solidifying its leadership in deep learning, while Facebook established its own AI lab, FAIR, to compete in the space [41][43]. - The acquisition of DeepMind by Google in 2014 expanded its AI capabilities but also led to internal conflicts between DeepMind and Google Brain [56][57]. Group 4: Emergence of OpenAI and Market Dynamics - OpenAI was founded in 2015 with a mission to promote and develop friendly AI, attracting talent from Google and other tech giants [66][68]. - The launch of ChatGPT in late 2022 marked a pivotal moment in the AI landscape, rapidly gaining users and prompting a competitive response from Google [97][99]. - Google's response included the rushed launch of Bard, which faced criticism and highlighted the challenges of adapting to disruptive innovations [102][103]. Group 5: Future Directions and Challenges - Google is now focusing on the Gemini project, aiming to unify its AI efforts and leverage its extensive resources to compete effectively in the evolving AI landscape [105][106]. - The competitive dynamics in the AI industry are shifting, with emerging players in China and the ongoing evolution of established companies like OpenAI and Meta [109][110].

袁进辉坦言，过去的屡败屡战都是自找的。在被光年之外并购前，OneFlow始终挣扎在生死线边缘，最 困难的时候选择了降薪、减员。他后来补充说，整个OneFlow六七年都是这样的状态。 文 | 施嘉翔 编辑 | 刘旌 袁进辉和OneFlow一度是ChatGPT爆发后最走运的那个： 在王慧文发布"英雄帖"后，OneFlow被光年之 外并购，这家成立6年的公司一跃成为牌桌上资源最好的团队。 但仅仅两个月后，一切又戏剧化地烟消 云散。 在聚光灯下经历跌宕起伏的四个月后，袁进辉和OneFlow团队决定重新创业。从有全行业最充沛的资 源，到突然消失，而其他玩家已经身处高位，心理冲击毋庸置疑。但袁进辉说，若仍想留在牌桌上，再 创业是唯一选择。 原因在于，王慧文生病的消息传出后，有公司冲到光年之外办公室挖人，OneFlow骨干成员也接到过超 千万的年薪，但没有一家公司能同时留住所有人。创业是避免团队被肢解的唯一方法。 在光年之外以前，袁进辉的创业同样起伏。和大部分公司不同，OneFlow成立之初就像实验室，前四年 没有任何收入，依赖外界投入。他们在做的事情在当时听起来匪夷所思：颠覆巨头重金投入的深度学习 系统框架（ 比如Goo ...

Core Insights - The article discusses the evolution of deep learning models from millions to billions of parameters, highlighting the lack of systematic understanding of their effectiveness [2] - A key focus is on the concept of parameter space symmetry, which refers to the existence of multiple parameter configurations that yield the same model function, complicating optimization and generalization analysis [4][6] Group 1: Parameter Space Symmetry - Parameter space symmetry allows different parameter combinations to produce identical outputs, exemplified by the interchange of neurons in hidden layers [4][6] - This symmetry is mathematically defined as transformations that keep the loss function invariant, forming a group that defines equivalent orbits in parameter space [6] Group 2: Types of Symmetry - In addition to discrete symmetries, most neural network architectures exhibit continuous symmetries, such as scaling and linear transformations, which maintain function invariance [8] - Complex architectures like Transformers combine various symmetries from their components, including multi-head attention mechanisms [8] Group 3: Impact on Loss Landscape - Symmetry creates a complex yet structured optimization space, where continuous symmetries can stretch isolated minima into flat manifolds, affecting the interpretation of generalization metrics [10] - Observed phenomena like "mode connectivity," where independently trained models can connect through low-loss paths, are partially attributed to continuous symmetries [10] Group 4: Optimization Methods - The presence of symmetry leads to the phenomenon of "equal loss, different gradients," suggesting new algorithmic possibilities for optimization methods that seek better gradient points within equivalent orbits [15][19] - Some optimization strategies leverage symmetry as a degree of freedom, while others aim to reduce it as redundancy, indicating its importance in algorithm design [19] Group 5: Learning Dynamics - Continuous symmetries correspond to conserved quantities, which remain constant during training, revealing insights into the stability of the training process and the implicit bias of optimization [21][23] - The structure of parameter space symmetry influences the statistical distribution of learning trajectories and outcomes [23] Group 6: Connections Across Spaces - Parameter space symmetry is interconnected with data space and internal representation space, where model parameters often reflect the symmetry present in the data distribution [27][28] - Emerging directions like Weight Space Learning utilize symmetry as a new data structure, facilitating the analysis and generation of model properties [28][29] Group 7: Future Directions - The widespread existence of parameter space symmetry offers a new mathematical language for deep learning, linking complex behaviors of models with established tools from group theory and geometry [30] - This perspective is influencing various practical fields, from optimization acceleration to model fusion and new model design, transforming theoretical concepts into actionable algorithmic principles [30]

Core Insights - OpenAI has completed a significant restructuring, laying the groundwork for an initial public offering (IPO) that could raise substantial funds to support its extensive computational and research initiatives [2][6][47] - The new structure maintains control by a nonprofit board, OpenAI Foundation, which holds 26% of the for-profit entity, OpenAI Group PBC, valued at approximately $1.3 trillion based on a recent $500 billion valuation [2][5] - Microsoft has become the largest shareholder with about 27% ownership, valued at approximately $135 billion, and has entered into a new agreement with OpenAI for additional Azure cloud services [5][6] Financial Structure - The restructuring converts previous investments into common equity, removing potential profit caps for investors while ensuring nonprofit oversight on critical governance matters [2][5] - OpenAI employees and early investors collectively hold about 26% equity, valued at around $1.3 billion, while future financing rounds will allocate 15% and 4% equity to new investors [5] - The company anticipates cash consumption exceeding $115 billion by 2029, driving the need for public market fundraising [6] Strategic Goals - OpenAI's CEO, Sam Altman, emphasized the importance of the restructuring as a pivotal event of the year, transitioning from a limited liability company (LLC) to a public benefit corporation (PBC) while retaining nonprofit control [7][25] - The company aims to develop a high-level AI research assistant by September next year and achieve fully automated AI researchers by March 2028 [12][36] - OpenAI's research focuses on deep learning technologies, with expectations of achieving superintelligence within a decade [10][11] Infrastructure and Product Development - OpenAI plans to build a robust infrastructure, committing over $1.4 trillion in financial responsibilities, with an initial focus on AI applications in healthcare [23][27] - The company is transitioning towards a platform model, allowing third-party developers to create applications based on OpenAI's technology [21][23] - OpenAI aims to produce 1 gigawatt of computing capacity weekly, with a target cost of approximately $20 billion per gigawatt over a five-year equipment lifecycle [24] Safety and Ethical Considerations - OpenAI is prioritizing safety and alignment in AI development, focusing on value alignment, goal alignment, reliability, adversarial robustness, and system safety [13][17] - The organization is committed to building an ecosystem for AI resilience, addressing potential risks associated with advanced AI technologies [28][30] - Altman highlighted the need for strong privacy protections as AI becomes a foundational platform in people's lives [23]

Core Viewpoint - OpenAI has completed a significant restructuring to transition from a non-profit organization to a profit-oriented entity while maintaining a commitment to its original mission of benefiting humanity through AGI development [4][12][13]. Summary by Sections Restructuring Announcement - OpenAI announced its restructuring plan, which aims to release its limited-profit subsidiary from direct control of the non-profit parent organization, allowing for stock issuance and potential IPO [4][12]. Historical Context - OpenAI was founded in 2015 as a non-profit with the goal of ensuring AGI benefits all of humanity, emphasizing long-term research without profit constraints [5][6]. - The organization faced funding challenges as the costs of developing AGI grew, leading to the establishment of a "capped-profit" subsidiary in 2019 to attract investment while limiting returns to investors [6][8]. New Structure - The new structure includes the OpenAI Foundation, which holds 26% of the equity and retains control, and the OpenAI Group PBC, which is a public benefit corporation eligible for IPO [13]. - Microsoft holds approximately 27% of the new structure, with the remaining shares distributed among employees and early investors, pushing OpenAI's valuation to around $500 billion [15][13]. Market Reaction - Following the restructuring announcement, Microsoft's stock rose by 4%, contributing to a market capitalization exceeding $4 trillion [14]. Future Goals - OpenAI aims to develop an AI assistant capable of conducting research by September 2026 and a fully automated AI researcher by March 2028 [20]. - The organization is focused on accelerating scientific discovery as a long-term impact of AGI [20]. Q&A Highlights - OpenAI addressed various user concerns during its first Q&A session, including the balance between user safety and freedom, the future of its models, and the potential for AI to automate cognitive tasks [24][30][44]. - The company acknowledged the need for age verification to enhance user autonomy while ensuring safety [26][30]. Financial Projections - OpenAI anticipates needing annual revenues in the range of several hundred billion dollars to support its projected $1.4 trillion investment needs [47].

Core Insights - The 2025 World Forum on Science and Technology Development opened in Beijing, highlighting the role of deep learning technology in AI-driven industrial digital transformation [1] - Deep learning is identified as a key technology that has significantly advanced AI capabilities, providing a foundation for the emergence of large models [1][2] Deep Learning Platform - The deep learning platform connects hardware (chips) with large models and applications, essential for AI development, training, inference deployment, and industrial implementation [2] - Baidu's PaddlePaddle serves as an industry-grade open-source deep learning platform, supporting the evolution of the Wenxin large model through a comprehensive framework, model library, and development tools [2][3] - PaddlePaddle has adapted to over 60 chip series and created more than 1.1 million models, showcasing its extensive capabilities [3] Model Performance and Optimization - The collaboration between PaddlePaddle and Wenxin has led to significant performance improvements, including a 47% increase in pre-training MFU for the ERNIE-4.5-300B-A47B model [3] - The model achieves high throughput performance, processing 57K tokens per second for input and 29K tokens per second for output under specific latency conditions [3] Industry Applications - The Wenxin large model has been recognized for its performance in various benchmarks, ranking first in domestic evaluations for multimodal and precise instruction-following tasks [4] - Baidu's deep learning platform is crucial for maximizing the impact of large models across industries, enhancing efficiency and decision-making capabilities [4] Specific Use Cases - In smart manufacturing, CRRC Group utilized PaddlePaddle to reduce high-speed train design simulation time from days to seconds [5] - In smart healthcare, AI optimizes patient experience and doctor efficiency through various stages of medical processes [5] - In smart energy, Baidu's technology has enabled comprehensive monitoring and intelligent decision-making for over 600 plants and 90 sections in the Guangxi power grid [5] Digital Human Technology - Baidu's digital human technology integrates multiple innovative techniques, resulting in highly interactive and realistic digital personas [6] - The commercial value of digital humans is evident, with over 100,000 digital anchors created, achieving a 31% increase in live streaming conversion rates and an 80% reduction in broadcasting costs [6] - The application of digital humans has outperformed real individuals in online performance, setting new records in sales during live broadcasts [6] Developer Engagement - The number of developers using PaddlePaddle and Wenxin has surpassed 23.33 million, serving over 760,000 enterprises [6]