Core Insights - The article highlights the advancements in 3D Gaussian Splatting (3DGS) technology, particularly its application in autonomous driving, and emphasizes the need for structured learning pathways in this rapidly evolving field [2][4]. Group 1: 3DGS Technology and Developments - Tesla's introduction of 3D Gaussian Splatting at ICCV has garnered significant attention, indicating a shift towards feed-forward GS algorithms in the industry [2]. - The rapid iteration of 3DGS technology includes static reconstruction (3DGS), dynamic reconstruction (4DGS), and surface reconstruction (2DGS), showcasing the need for effective learning resources [4]. Group 2: Course Offering - A comprehensive course titled "3DGS Theory and Algorithm Practical Tutorial" has been developed to provide a structured learning roadmap for newcomers, covering essential theories and practical applications [4]. - The course is designed to help participants understand point cloud processing, deep learning, real-time rendering, and coding practices, with a focus on hands-on experience [4]. Group 3: Course Structure - The course consists of six chapters, starting with foundational knowledge in computer graphics and progressing to advanced topics such as feed-forward 3DGS and its applications in autonomous driving [8][9][10][11][12]. - Each chapter includes practical assignments and discussions to enhance understanding and application of the concepts learned [8][9][10][11][12]. Group 4: Target Audience and Prerequisites - The course is aimed at individuals with a background in computer graphics, visual reconstruction, and programming, particularly those interested in pursuing careers in the 3DGS field [17]. - Participants are expected to have a foundational understanding of probability, linear algebra, and programming languages such as Python and PyTorch [17].

Core Insights - A research team from MIT, the University of Waterloo, and Samsung Electronics has developed a new method to increase transistor density on chips by stacking additional layers of transistors on existing circuits, which could significantly enhance chip performance and energy efficiency [2][4][5]. Group 1: New Manufacturing Method - The new method involves adding a layer of micro-switches on top of completed chips, similar to traditional chip stacking techniques, to increase the number of transistors integrated into a single chip [2]. - The research team utilized a 2-nanometer thick layer of amorphous indium oxide to construct additional transistors without damaging the sensitive front-end components during the manufacturing process [3][6]. Group 2: Energy Efficiency and Performance - This innovative approach allows for the integration of logic devices and memory components into a compact structure, reducing energy waste and improving computational speed [4][5]. - The new transistors exhibit a switching speed of just 10 nanoseconds, with significantly lower voltage requirements compared to existing devices, leading to reduced power consumption [6]. Group 3: Future Implications - The research indicates that if future processors can utilize both this new technology and traditional chip stacking methods, the limits of transistor density could be greatly surpassed, countering the notion that Moore's Law is reaching its end [3][4]. - The team aims to further integrate these backend transistors into single circuits and enhance their performance, exploring the physical properties of ferroelectric hafnium zirconium oxide for potential new applications [7].

Core Insights - The article discusses the transformative potential of Artificial Intelligence (AI) as the fourth industrial revolution, emphasizing its role in enhancing productivity and reducing costs across various sectors [1][5]. Group 1: AI Development Drivers - AI is driven by the need to improve efficiency and reduce costs, addressing pain points in consumer-related scenarios such as entertainment, travel, and health [3]. - The application of AI in consumer sectors includes labor replacement and productivity enhancement through technologies like voice recognition and intelligent customer service [3]. - In the business sector, AI is widely adopted in finance, public safety, and healthcare, reflecting a strong demand for efficiency improvements [3]. Group 2: Historical Context and Evolution - AI is positioned as the fourth productivity revolution, following the steam, electrical, and information technology revolutions, with significant historical milestones marking its development [5]. - The evolution of AI has seen three major waves of growth, each driven by breakthroughs in underlying algorithms, with the current wave characterized by deep learning advancements [8][12]. - The first wave of AI in the 1950s was limited by computational performance, while the second wave in the 1980s faced challenges due to the high costs of expert systems [9][11]. Group 3: AI Industry Structure - The AI industry can be segmented into three layers: foundational support (hardware and data), technology (algorithm development), and application (commercial solutions) [6][7]. - Major players in the foundational layer include international tech giants like Nvidia and Intel, while the technology layer features companies like Google and IBM focusing on specific AI applications [6][7]. - The application layer is where AI technologies are commercialized, with a relatively low entry barrier due to the global open-source community [6]. Group 4: Current AI Landscape - The current state of AI is classified as "weak AI," focusing on specific tasks such as speech and image recognition, with significant performance exceeding human capabilities in certain areas [30][33]. - AI's impact on global GDP is projected to be substantial, with estimates suggesting a 14% increase, translating to approximately $15.7 trillion in growth [37]. - The rapid advancement of deep learning algorithms and the availability of vast datasets are expected to drive widespread AI application across various industries [38][39]. Group 5: Future Opportunities - The article highlights the potential for AI to revolutionize content generation and distribution, particularly through platforms like TikTok and Douyin, which utilize AI-driven recommendation systems [46][52]. - The emergence of generative AI (AIGC) is seen as a significant opportunity, with advancements enabling the creation of diverse content types, including text, images, and videos [54][61]. - The integration of AI into various sectors is anticipated to accelerate, driven by technological advancements and supportive policies from governments [44][45].

公开资料显示，苏箐曾担任华为车 BU 智能驾驶产品部部长，负责华为自动驾驶系统方案。2022 年 1 月，苏箐正式从华为离职。同年 10 月，苏箐加入地平 线。 苏菁称，关于特斯拉 FSD V12 到底是不是最强的问题，业内争议很大，但这个问题不重要，重要的是 FSD V12 证明了一段式端到端技术的可行性，推动智 驾技术范式从规则驱动转向数据驱动。他认为，一段式端到端在智驾行业的普及将带来两大趋势的行业演进。 一是智驾系统会在未来几年内越来越"类人"，这将使 L2 级辅助驾驶迎来巨大的发展红利期，城区辅助驾驶将逐步普及到 10 万元级别车型。 二是 L2 和 L4 级别的智驾方法论统一，同样的开发范式，不仅能提升 L2 辅助驾驶体验，同时也能以更低的部署成本和几乎无限制部署区域扩张，落 地一个 L4 系统（Robotaxi）。 2024年,以FSD V12成熟为标准, 智能驾驶迎来内在底层技术 范式与外部用户感知体验的一次 重构。其意义,堪比核能从理论 迈入工程。 苏 等 地平线副总裁兼首席架构师 2025 地平线技术生态大会 HORIZON 75 TOGETHER Z IT之家 12 月 12 日消息，据 ...

Core Insights - The article discusses the evolution of AI research and the transition from scaling to a renewed focus on innovative research methods, emphasizing the importance of "taste" in research and the potential for breakthroughs in AI learning mechanisms [10][12][16]. Group 1: Transition of AI Research - The AI development is shifting from a scaling era (2020-2025) back to a research-focused era, as the scaling laws of pre-training are becoming ineffective due to limited data [17]. - The future of AI is expected to involve new algorithms rather than just increasing computational power [17]. Group 2: SSI's Strategy - Safe Superintelligence Inc. (SSI) aims to develop superintelligence without intermediate products, focusing solely on research rather than market competition [12]. - Ilya Sutskever, co-founder of SSI, believes that the company’s funding of $3 billion is entirely directed towards research, unlike larger companies that allocate funds to user services and sales teams [13]. Group 3: Research Methodology - Ilya emphasizes the importance of a "Value Function" in AI learning, suggesting that current reinforcement learning (RL) methods are inefficient and may hinder the model's capabilities [16][20]. - He proposes that future breakthroughs in AI will come from enabling models to make intuitive judgments during the learning process [19]. Group 4: Emotional Intelligence in AI - Ilya argues that emotions serve as a crucial decision-making tool for humans, and AI currently lacks this capability, which may be essential for achieving AGI [22]. - He suggests that empathy could be a fundamental aspect of AI development, allowing AI to understand and care for sentient life [24]. Group 5: Market Dynamics - The future AI market is expected to be competitive and specialized, with companies focusing on niche areas rather than a single entity dominating superintelligence [28]. - This specialization will create high barriers to entry for new competitors, similar to ecological balances in nature [28].

【导读】目标疯狂，一路偶然！奥特曼回顾OpenAI十年，坦承吃到了时代的红利。 今天一睁眼，大家都被OpenAI十周年的生日祝福刷屏了。 转眼间，这个改变了全世界的AI初创，如今已经成为巨头。 一位OpenAI的老员工，晒出自己在2019年在OpenAI第一天上班的照片 凌晨，和GPT-5.2一起来临的，还有OpenAI的十周年。 OpenAI发布了一支短片，配文只有两个词：「10年」。 这支短片，其实讲的不是产品，而是一种信念。 画面从OpenAI注册那天开始： 一群技术宅，挤在厨房里，讨论一个听起来像科幻小说中的目标：AGI。 但这支十周年视频，也留下了一个明显的空白。 镜头里，没有Ilya和Mira的身影。 而他们，恰恰是为 OpenAI 打下前五年地基的人。 有些人，奠基了历史，却没出现在纪念片里 如今，OpenAI估值800亿美元、超过1000名员工，打造了全球用户最多的大语言模型。 但要回顾OpenAI的10年，奥特曼绝对是主角。 奥特曼亲自发文，庆祝十周年 十年前，AI连猫和狗都分不清。 但OpenAI的创始人相信，深度学习能走得更远，相信它可能成为人类的一项重大胜利。 然而，过去的11位联合创 ...

Core Insights - The article highlights the significance of the "Test of Time Award" received by the paper "Faster R-CNN," co-authored by renowned researchers, marking its impact on the field of computer vision since its publication in 2015 [1][5][25] - The presentation by He Kaiming at NeurIPS 2025 summarizes the evolution of visual object detection over the past 30 years, showcasing key milestones and influential works that have shaped the field [6][31] Historical Development - The early attempts at face detection in the 1990s relied on handcrafted features and statistical methods, which were limited in adaptability and speed [12] - The introduction of AlexNet in 2012 demonstrated the superior feature extraction capabilities of deep learning, paving the way for its application in object detection [15] - The R-CNN model, proposed in 2014, revolutionized object detection by integrating CNNs for feature extraction and classification, although it initially faced computational challenges [17][18] Technological Advancements - The development of Faster R-CNN in 2015 addressed the speed bottleneck by introducing the Region Proposal Network (RPN), allowing for end-to-end real-time detection [25] - Subsequent innovations, such as YOLO and SSD in 2016, further enhanced detection speed by enabling direct output of object locations and categories [32] - The introduction of Mask R-CNN in 2017 added instance segmentation capabilities, while DETR in 2020 redefined detection using Transformer architecture [32][34] Future Directions - The article concludes with reflections on the ongoing exploration in computer vision, emphasizing the need for innovative models to replace outdated components as bottlenecks arise [35][36]

Core Insights - The autonomous driving industry is expected to transition from rapid paradigm shifts to a phase of extreme optimization over the next three years, as stated by a veteran in the field [2][3] - The release of FSD V12 in 2024 is seen as a watershed moment for the industry, marking a significant technological breakthrough that could resolve long-standing bottlenecks [2][3] - Current deep learning technologies are showing signs of reaching their limits, and without breakthroughs in AGI theory, the industry may face a prolonged period of optimization rather than innovation [3][4] Industry Trends - The FSD V12's end-to-end architecture breaks existing barriers by extending deep learning applications from perception to decision-making, completing a technological revolution [3] - The paradigm shift allows for shared development frameworks and sensor configurations between L2 and L4 systems, enhancing collaboration and efficiency [3] - The industry is advised to focus on maximizing the potential of existing technologies, with an emphasis on improving chip performance and model capacity [4] Strategic Directions - The company plans to achieve a tenfold increase in computing power for each generation of AD products, supporting a tenfold scale of system evolution [3] - There is a focus on making L2 systems accessible to a broader market, targeting a price point that allows for wider adoption [4] - The ultimate goal remains to create machines that can replace human drivers, emphasizing the importance of endurance and precision in the industry’s long-term efforts [4]

Core Insights - The article discusses the rapid advancements in 3D Generative Synthesis (3DGS) technology, highlighting its applications in various fields such as 3D modeling, virtual reality, and autonomous driving simulation [2][4] - A comprehensive learning roadmap for 3DGS has been developed to assist newcomers in mastering both theoretical and practical aspects of the technology [4][6] Group 1: 3DGS Technology Overview - The core goal of new perspective synthesis in machine vision is to create 3D models from images or videos that can be processed by computers, leading to numerous applications [2] - The evolution of 3DGS technology has seen significant improvements, including static reconstruction (3DGS), dynamic reconstruction (4DGS), and surface reconstruction (2DGS) [4] - The introduction of feed-forward 3DGS has addressed the inefficiencies of per-scene optimization methods, making the technology more accessible [4][14] Group 2: Course Structure and Content - The course titled "3DGS Theory and Algorithm Practical Tutorial" covers detailed explanations of 2DGS, 3DGS, and 4DGS, along with important research topics in the field [6] - The course is structured into six chapters, starting from foundational knowledge in computer graphics to advanced topics like feed-forward 3DGS [10][11][14] - Each chapter includes practical assignments and discussions to enhance understanding and application of the concepts learned [10][15] Group 3: Target Audience and Prerequisites - The course is designed for individuals with a background in computer graphics, visual reconstruction, and programming, particularly in Python and PyTorch [19] - Participants are expected to have a GPU with a recommended computing power of 4090 or higher to effectively engage with the course material [19] - The course aims to benefit those seeking internships, campus recruitment, or job opportunities in the field of 3DGS [19]

Core Insights - The discussion highlights the transformative impact of artificial intelligence (AI) and the role of NVIDIA in driving this technological revolution, emphasizing the importance of GPUs in various applications from gaming to modern data centers [2][10]. Group 1: AI and Technological Competition - The conversation underscores that the world is in a significant technological race, particularly in AI, where the first to reach advanced capabilities will gain substantial advantages [11][12]. - Historical context is provided, indicating that the U.S. has always been in a technological competition since the Industrial Revolution, with AI being the latest frontier [12][13]. Group 2: Energy and Manufacturing - The importance of energy growth and domestic manufacturing is emphasized as critical for national security and economic prosperity, with a call for revitalizing U.S. manufacturing capabilities [8][9]. - The discussion points out that without energy growth, industrial growth and job creation would be severely hindered, linking energy policies directly to advancements in AI and technology [9][10]. Group 3: AI Development and Safety - Concerns about the risks associated with AI are acknowledged, particularly regarding its potential military applications and ethical implications [19][20]. - The conversation suggests that AI's development will be gradual rather than sudden, with a focus on enhancing safety and reliability in AI systems [14][15]. Group 4: Future of AI and Knowledge Generation - The potential for AI to generate a significant portion of knowledge in the future is discussed, with predictions that AI could produce up to 90% of knowledge within a few years [41][42]. - The necessity for continuous verification of AI-generated information is highlighted, stressing the importance of ensuring accuracy and reliability in AI outputs [41][42]. Group 5: Cybersecurity and Collaboration - The dialogue emphasizes the collaborative nature of cybersecurity, where companies share information and best practices to combat threats collectively [23][24]. - The need for a unified approach to cybersecurity in the face of evolving threats is reiterated, suggesting that cooperation is essential for effective defense [23][24].