Workflow
量子位
icon
Search documents
ImageNet作者苏昊被曝任教复旦
量子位· 2025-10-10 03:52
Core Viewpoint - The article discusses the potential appointment of Hao Su, a prominent figure in embodied intelligence and computer vision, to Fudan University, highlighting his significant contributions to the field and his entrepreneurial ventures in robotics and simulation [1][49][51]. Group 1: Hao Su's Academic and Research Background - Hao Su is an associate professor at the University of California, San Diego (UCSD), specializing in computer vision, graphics, embodied intelligence, and robotics [14][49]. - He was involved in the creation of ImageNet and has led foundational projects such as ShapeNet, PointNet, and SAPIEN, which have significantly advanced the fields of 2D and 3D vision [4][30][34]. - Su's research has evolved from natural language processing to computer vision and then to 3D vision, culminating in the development of large-scale datasets and models that have transformed the landscape of artificial intelligence [22][30][34]. Group 2: Contributions to Robotics and Simulation - In 2020, Su launched SAPIEN, the first simulator focused on generalizable robotic operations, and later developed the ManiSkill platform for training robotic skills [35][41]. - His company, Hillbot, co-founded in 2024, aims to leverage high-fidelity simulation for robotics, with products like Hillbot Alpha designed for complex environments [43][45]. - Hillbot has partnered with Nvidia to generate high-quality training data, indicating a strong focus on enhancing robotic capabilities through advanced simulation techniques [47]. Group 3: Potential Move to Fudan University - There are rumors that Su will join Fudan University, which may invest in his company Hillbot and potentially appoint him to dual roles at various research institutes [51][52]. - Fudan University has established a credible embodied intelligence research institute, offering competitive salaries and performance-based incentives, which could attract top talent like Su [55][57].
230个大模型在婴幼儿认知题上集体翻车!揭秘多模态大模型的核心知识缺陷
量子位· 2025-10-10 01:03
Core Insights - The article highlights that while most AI models perform well on complex tasks, they struggle significantly with basic cognitive abilities that humans develop from a young age [1][4]. Core Cognition Benchmark - Researchers created the CoreCognition benchmark, which includes 1503 classic developmental psychology tests covering 12 core cognitive abilities that emerge in early childhood [2][9]. - The benchmark aims to systematically test models on their understanding of fundamental cognitive concepts such as object permanence and intuitive physics [5][9]. Model Performance - A comparison of 230 mainstream models revealed a "core knowledge blind spot," with many models showing significant deficits in basic cognitive abilities, often lagging behind human performance by double-digit percentages [3][4][16]. - The study found that lower-level cognitive abilities (e.g., boundary perception, continuity) are significantly weaker in models compared to higher-level abilities (e.g., intentional understanding, mechanical reasoning) [16][18]. Key Findings - The research identified five key findings regarding the cognitive capabilities of models: 1. Models exhibit systematic shortcomings in foundational "core knowledge" compared to human cognitive development [16]. 2. There is a weak correlation between lower-level abilities and higher-level reasoning, indicating a lack of scaffolding in cognitive development [18]. 3. Core abilities are positively correlated with performance on public benchmarks, suggesting that stronger core knowledge leads to better task performance [20]. 4. Increasing model size does not significantly improve lower-level cognitive abilities, with some abilities even deteriorating as model size increases [22]. 5. Concept Hacking experiments showed that larger models do not necessarily perform better, indicating that mere scaling does not eliminate reliance on shortcuts [24]. Cognitive Instruction and Model Understanding - Cognitive instructions can provide short-term gains in performance, but they do not address the underlying gaps in foundational knowledge [27][29]. - The study suggests that true intelligence relies on understanding the most basic rules of the world, rather than just increasing model parameters [31][32]. Recommendations - The article advocates for a shift in focus from merely scaling models to ensuring that foundational cognitive knowledge is solidified first, emphasizing that core knowledge is multiplicative rather than additive [33][34].
英特尔发布“2纳米级”工艺CPU,你的AI PC性能即将暴涨50%
量子位· 2025-10-10 01:03
Core Insights - Intel's Panther Lake processors have officially entered mass production, showcasing the new 18A process technology and significant performance improvements across CPU, graphics, and AI capabilities [1][29][25] Product Development - The previous generation's design, which integrated memory into the processor, has been abandoned to streamline the product line, creating a versatile platform that spans from ultrabooks to high-performance gaming laptops [2][5] - Panther Lake features a multi-tile architecture that optimizes performance and efficiency, integrating various functional modules into a single package [9][12] Performance Metrics - Compared to the previous generation, Panther Lake boasts over a 50% increase in multi-core and graphics performance, while overall power consumption has decreased by 30% [3] - The new architecture includes advanced CPU cores, with the "Cougar Cove" performance core showing a 10% improvement in single-threaded performance over the previous generation [15] Memory and Cache Enhancements - Panther Lake supports LPDDR5x memory up to 9600 MT/s (96GB) and DDR5 memory up to 7200 MT/s (128GB), exceeding many desktop platform specifications [12][14] - An 8MB memory-side cache has been added, reducing DRAM access latency by 30% and achieving a 95% data hit rate, which is crucial for AI applications [19] Graphics and AI Capabilities - The new Xe3 "Celestial" graphics architecture offers over a 50% performance increase compared to the previous generation [20] - Panther Lake includes a fifth-generation NPU providing 50 TOPS of dedicated AI performance, with the total platform AI capability reaching 180 TOPS, significantly surpassing Microsoft Copilot+ PC requirements [25] Manufacturing Innovations - Panther Lake is the first product to utilize Intel's 18A process technology, which is critical for Intel's return to semiconductor manufacturing leadership, offering a 15% performance per watt improvement and a 30% increase in chip density [29][30] - Key innovations in the 18A process include the RibbonFET transistor architecture and PowerVia back-side power delivery technology, enhancing efficiency and performance [32][34] Future Outlook - Panther Lake is expected to be officially launched at CES in January 2026, with the first laptops featuring the new processors to follow shortly after [28] - Concurrently, Intel has previewed the next-generation server processor, Clearwater Forest, also based on the 18A process, anticipated for release in the first half of 2026 [39]
74岁“酒鬼”教授终于拿诺贝尔化学奖了
量子位· 2025-10-09 09:34
Core Viewpoint - The Nobel Prize in Chemistry this year was awarded to scientists for their significant contributions to Metal-Organic Frameworks (MOFs), which are materials capable of storing large amounts of gas in a small volume, akin to Hermione's bag in Harry Potter [1] Group 1: Awarded Scientists and Their Contributions - One of the laureates, Kitagawa Jun, is known as the father of Porous Coordination Polymers (PCPs) and demonstrated that gases can flow in and out of MOF structures, predicting their potential for creating flexible materials [3] - Richard Robson was the first to discover the concept of MOFs, inspired by the diamond structure, and designed an ordered pyramid crystal structure [10] - Omar M. Yaghi formally introduced the term MOF and developed stable linkers made from carboxylate groups, which can be modified to impart new desirable properties [14] Group 2: Significance and Potential of MOFs - The Nobel Committee highlighted the immense potential of MOFs to provide unprecedented opportunities for custom materials with new functionalities [7] - MOFs possess a unique porous structure that allows for various applications, such as collecting water from desert air, capturing carbon dioxide, and storing toxic gases [10] - A report from Nature indicated that MOFs are likely one of the most researched materials in the past 20 years, with over 100,000 academic papers related to them, resulting in thousands of different functional MOFs [16] Group 3: Commercialization and Challenges - Despite the extensive research, only a few MOFs have been commercialized, such as CALF-20 by Canadian company Svante, which is used to scale up the removal of CO2 from cement production waste [17] - The main challenge lies in finding the correct material combinations, but once identified, the possibilities are vast [18] Group 4: Personal Insights of Kitagawa Jun - Kitagawa Jun's choice of chemistry as a career was humorously linked to his ability to distinguish between genuine and fake alcohol [4][25] - He is known for his unconventional approach to mentoring students, often encouraging them to engage socially over drinks to foster communication and confidence [30][29] - His unique perspective and intuition in research have been pivotal in his success, emphasizing the importance of luck, patience, and perseverance in scientific discovery [46][47]
库克被曝让贤:接棒乔布斯任苹果CEO已14年
量子位· 2025-10-09 09:34
Core Viewpoint - Apple is reportedly planning a significant leadership change, with Tim Cook potentially stepping down as CEO and being succeeded by John Ternus, the current Senior Vice President of Hardware Engineering [1][4][6]. Group 1: Leadership Transition - Tim Cook may transition to the role of Chairman, following a path similar to that of Jeff Bezos and Bill Gates [3]. - The leadership shake-up is described as the largest in over a decade, with several key executives considering their future roles within the company [4][5]. - John Ternus, aged 50, is positioned as a strong candidate for the CEO role, having gained significant trust and influence within Apple [6][24]. Group 2: Tim Cook's Tenure - Tim Cook took over as CEO in 2011, inheriting a company with a strong product lineup, including the iPhone and iPad, which saw sales growth of approximately 200% from 2011 to 2016 [11][12]. - Under Cook's leadership, Apple became the first company to reach a market valuation of $3 trillion in January 2022 [16]. - Cook shifted focus towards higher-margin services, launching products like Apple Music, Apple TV+, and iCloud, which helped build a robust ecosystem [13][14]. Group 3: Challenges Faced - As Apple enters the AI era, Cook's conservative approach has drawn criticism, with many believing the company is lagging behind in AI advancements [17][18][20]. - The once-successful strategies under Cook are now seen as potential obstacles for Apple's future growth in the rapidly evolving tech landscape [19][40]. Group 4: John Ternus's Profile - John Ternus has been with Apple since 2001 and has played a crucial role in the development of key products, including the iPad and iPhone [26][28]. - Ternus is recognized for his collaborative approach, effectively bridging hardware and software teams, and is noted for his ability to manage large projects hands-on [35][36]. - His recent public appearances and interviews have showcased his strategic vision and leadership qualities, positioning him as a strong candidate for the CEO role [29][33].
阿里亲身入局具身智能!Qwen内部组团,通义千问技术负责人带队
量子位· 2025-10-09 07:03
Core Viewpoint - Alibaba has established a new team focused on embodied intelligence, marking a significant step in its exploration of physical AI systems, following in the footsteps of companies like OpenAI and Google [2][3][5]. Group 1: Team Formation and Leadership - The Qwen team, under Alibaba, has formed a dedicated embodied intelligence squad, which is part of the core department responsible for the development of the Qwen series of large models [6][7]. - Justin Lin, the technical lead of Qwen, personally set up this team, indicating a hands-on approach to its development [10][11]. - Lin has a strong background in AI foundational models, having transitioned from natural language processing to large-scale pre-trained models and now leading the Qwen project [12][18][19]. Group 2: Strategic Direction and Investment - Alibaba has been strategically focused on embodied intelligence since 2024, investing in several companies in this field, including Zhi Ji Power and Xingdong Era [21][22]. - In September 2023, Alibaba Cloud led a $140 million financing round for a robotics company, marking its first direct investment in the embodied intelligence sector [23]. - The company aims to integrate AI large models with robotics and automation, as highlighted in the "Physical AI" initiative announced at the 2025 Cloud Summit [24][25]. Group 3: Technological Evolution and Future Outlook - The establishment of the embodied intelligence team signals Alibaba's shift towards applying AI in real-world scenarios, moving beyond purely virtual applications [27][30]. - The growth in model scale has enhanced AI's capabilities in abstract reasoning and task decomposition, enabling a transition from software simulations to real-world applications [28][29]. - Alibaba's CEO has projected that global AI investment will exceed $4 trillion in the next five years, emphasizing the company's commitment to advancing AI technology towards embodied intelligence and robotic applications [30][31].
OpenAI奥特曼认错:我天生不适合管理公司
量子位· 2025-10-09 07:03
Core Insights - OpenAI is pursuing three main goals: to become a personal AI subscription service, to build large-scale infrastructure, and to achieve a truly useful AGI (Artificial General Intelligence) [2][4][29] - The recent launch of Sora 2 and various investment collaborations, including partnerships with AMD and Nvidia, indicate a strategic shift towards aggressive infrastructure investment [1][29] Group 1: OpenAI's Strategic Goals - OpenAI aims to become a personal AI subscription service, necessitating the construction of vast infrastructure to support this vision [4][29] - The ultimate mission is to create AGI that is genuinely beneficial to humanity, which requires a multifaceted approach beyond traditional business models [4][8] - OpenAI's infrastructure is currently intended for internal use, with future possibilities for external applications remaining uncertain [5][29] Group 2: Sora's Role in AGI Development - Despite skepticism about Sora's relevance to AGI, OpenAI's CEO believes that developing a "truly outstanding world model" through Sora will be crucial for AGI [10][11] - The resources allocated to Sora are relatively small compared to OpenAI's overall computational capacity, emphasizing a balanced approach to innovation and research [13][29] - Sora is seen as a way to engage society with upcoming technological advancements, particularly in video models, which resonate more emotionally than text [16][29] Group 3: Future Interactions and AI Capabilities - OpenAI envisions future interaction interfaces that go beyond basic chat, incorporating real-time video rendering and context-aware hardware [19][21] - The concept of the Turing Test is evolving, with the new benchmark being AI's ability to conduct scientific research, which OpenAI anticipates will happen within two years [21][22] - OpenAI's confidence in its research roadmap and the economic value it can generate has led to a commitment to aggressive infrastructure investments [29][31] Group 4: Leadership and Management Philosophy - OpenAI's CEO acknowledges a preference for an investor role over management, citing challenges in handling organizational dynamics and operational details [41][42] - The transition from an investor to a CEO role has been described as both challenging and rewarding, providing insights into groundbreaking work in AI [41][43] - The future of AI development is closely tied to energy availability, with a call for more efficient energy solutions to support AI advancements [44]
黄仁勋回应AMD“送股”OpenAI:很高明的交易,OpenAI现在还没钱给我付账
量子位· 2025-10-09 04:52
Core Insights - Huang Renxun expressed surprise at AMD's strategy of exchanging 10% of its shares for OpenAI orders, calling it clever [1][3] - He emphasized that NVIDIA's relationship with OpenAI is fundamentally different, as NVIDIA sells products directly to OpenAI rather than through equity exchange [4] - OpenAI currently lacks the funds to pay for its large orders, needing to raise capital through future revenue growth, equity, or debt financing [5][7] NVIDIA and OpenAI Transactions - NVIDIA has the opportunity to co-invest in OpenAI's future financing rounds, with Huang expressing regret for not investing more when OpenAI was founded [8] - NVIDIA announced plans to invest up to $100 billion in OpenAI, which will build systems requiring 10 gigawatts of power, equivalent to 4 to 5 million GPUs [12][13] - OpenAI has also agreed to purchase AMD chips, committing to a significant procurement of AMD's upcoming MI450 series [14] Circular Trading Dynamics - The transactions create a closed-loop funding mechanism where NVIDIA's investment flows to Oracle through OpenAI, which then purchases NVIDIA hardware [16] - The total value of AI computing transactions between OpenAI, NVIDIA, AMD, and Oracle has surpassed $1 trillion, while OpenAI is projected to struggle with positive cash flow until 2029 [16] Expansion of NVIDIA's AI Investments - NVIDIA confirmed its participation in a $20 billion funding round for Musk's xAI, with plans to invest $2 billion [17] - The investment will utilize a special purpose vehicle (SPV) structure, with funds allocated for purchasing NVIDIA processors [18] - NVIDIA has also invested in CoreWeave, holding a 7% stake, and is actively involved in numerous AI venture capital transactions [19][20] Market Concerns - The intricate web of transactions has raised concerns about a potential AI bubble, with analysts warning that these deals could signal early warning signs if a bubble were to burst [20] - NVIDIA representatives clarified that the company does not require its invested companies to use NVIDIA technology [21][22]
2025人工智能年度评选启动!3大维度5类奖项,正在寻找AI+时代领航者
量子位· 2025-10-09 04:52
组委会 发自 凹非寺 量子位|公众号 QbitAI 为了让更多从业者感受智能浪潮的跃迁,也为了给予更多同行同路人掌声与鼓舞,我们将正式启动 「2025人工智能年度榜单」评选报名 。 这是量子位人工智能年度榜单的 第8年 。八年来,我们见证了技术的突破与落地,产业的融合与重塑,也见证了一批又一批推动时代前行的 企业、人物与产品。 在人工智能重新定义一切的时代里,智能技术已不再是单一工具,而是产业与社会协同进化的驱动力。我们期待通过这场年度评选,去发现并 致敬那些真正引领变革、开拓边界的探索者与实践者。 本次评选将从 企业 、 产品 、 人物 三大维度,设立五类奖项。欢迎企业踊跃报名! 让我们共同见证年度之星,点亮未来的方向。 企业榜 产品榜 人物榜 2025 人工智能年度 焦点人物 详细评选标准及报名方式如下。 2025 人工智能年度领航企业 2025 人工智能年度 领航企业 2025 人工智能年度 潜力创业公司 2025 人工智能年度 杰出产品 2025 人工智能年度 杰出解决方案 评选标准 : 1、注册地在中国,或主营业务主要面向中国市场; 2、主营业务属于人工智能及相关产业,或已将人工智能广泛应用于主营业 ...
备受Meta折磨,LeCun依旧猛发论文!新作:JEPAs不只学特征,还能精准感知数据密度
量子位· 2025-10-09 04:52
Core Insights - The article discusses a new research paper by Yann LeCun's team that reveals the hidden capability of the self-supervised model JEPAs (Joint Embedding Predictive Architecture) to learn data "density" [2][5][6] - This finding challenges the long-held belief that JEPAs only excel at feature extraction and are unrelated to data density [7] Group 1: Key Findings - JEPAs can autonomously learn the commonality of data samples during training, allowing them to assess the typicality of a sample without additional modifications [6][11] - The core discovery is that the anti-collapse mechanism enables precise learning of data density, which was previously underestimated [11][12] - The research highlights that when JEPAs output Gaussian embeddings, they must perceive data density through the Jacobian matrix, making the learning of data density an inherent result of the training process [11] Group 2: Practical Applications - The team introduced a key tool called JEPA-SCORE, which quantifies data density and scores the commonality of samples [14][15] - JEPA-SCORE is versatile and can be applied across various datasets and JEPAs architectures without requiring additional training [16][17] - Experiments demonstrated that JEPA-SCORE effectively identifies typical and rare samples across different datasets, confirming its reliability and general applicability [18] Group 3: Research Team - The research was a collaborative effort involving four core researchers from Meta's FAIR, including Randall Balestriero, Nicolas Ballas, and Michael Rabbat, each with significant backgrounds in AI and deep learning [26][28][30][32][34][36]