大公司： 美团宣布养老保险补贴将全国上线，预计可覆盖超百万名骑手 36氪获悉，7月17日下午，美团宣布今年内将在全国范围推广养老保险补贴，该社保计划预计将覆盖超 百万外卖骑手。美团相关负责人还介绍了多方面保障升级，从7月1日起，骑手"工伤险"，即"新职伤"已 扩展到17个省级行政区，为骑手提供意外伤害风险防护网；骑手安全方面，将继续升级事前预防、事中 干预和事后兜底的骑手安全体系，骑手安全跑单可获得现金奖励；骑手服务方面，未来三年新建1000 座"骑手之家"，由中华全国总工会合作建设落地，并面向全行业开放。 爱奇艺推出内容热度值"四大殿堂" 36氪获悉，爱奇艺基于平台内容热度值推出"剧集四大殿堂"系列，包括"荣誉殿堂"（内容热度值达 10000及以上的剧集）、"国民殿堂"（内容热度值达9500及以上的剧集）、"人气殿堂"（内容热度值达 9000及以上的剧集）、"佳片殿堂"（内容热度值达8000及以上的剧集）。据介绍，爱奇艺是国内第一家 推出内容热度值的视频平台。 哪吒汽车母公司预招募重整意向投资人 据阿里资产消息，合众新能源汽车股份有限公司（哪吒汽车母公司）重整意向投资人预招募通道正式开 启，自2025年7月 ...

一生要出片的中国人，出门旅行最大的快乐，不是买买买，而是拍拍拍。在社交媒体主导的消费时代，"出片"已从一种自发行为演变为用户的核心需求。 数据显示，小红书上"#出片"话题浏览量高达4.1亿，参与讨论的人超过139.6万；抖音上，同样话题的视频播放量也达到了12.1亿次。这股全民拍照热潮背 后，藏着怎样的消费心理？品牌又该怎么抓住机会呢？ | 全部三 用户 | 商品 | 话题 | 地点 | | 出片 | | | 8 | 搜索 | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | | 直播 | 音乐 | 团购 话题 | 图文 | 小程序 | 店镇 | | # 出片 | 4.1亿浏览 139.9万讨论 | | 去发布 | | | | | | | | | | | | | | #出片 | | 立即参与 | | | 井 出片圣地 | | | | | | 12.1亿次播放 | | | | | 4455.3万浏览 23.4万讨论 | | | 去发布 | | | | | | | | | | | | | | #出片神器 2.5亿次播放 | ...

Group 1 - The core idea of the article is that the group broadcasting (团播) industry is evolving rapidly, transitioning from a simplistic entertainment format to a more structured and professional model akin to idol groups, particularly influenced by the success of digital platforms and the influx of new talent [2][5][6] - By the second half of 2025, group broadcasting is expected to become a significant component of short video algorithms, reshaping the live streaming industry with more organized performances and audience engagement [2][6] - The revenue potential in group broadcasting is substantial, with top streamers reportedly earning over 10 million yuan (approximately 1.4 million USD) in a single event, attracting more newcomers to the field and increasing professionalism [5][6] Group 2 - The evolution of group broadcasting resembles a digital upgrade of the AKB48 idol training model, with a focus on fan-driven resource allocation and competition [8][10] - Unlike traditional idol groups, group broadcasting allows for immediate feedback and competition, with performance metrics updated every few minutes, enhancing the engagement between fans and streamers [10][11] - The selection of popular streamers often favors "ordinary" individuals, similar to the rise of AKB48's early members, indicating a trend towards authenticity and relatability in talent [10][11] Group 3 - Emotional connections between fans and streamers are deepened through interactive features, such as direct communication and personalized engagement, creating a stable viewing relationship [13][20] - Group broadcasting provides brands with new marketing opportunities, leveraging the extensive reach and positive image of idol-like performers compared to traditional entertainment formats [14][15] - Brands are increasingly forming their own group broadcasting teams, indicating a shift towards integrated marketing strategies that utilize live streaming as a promotional tool [17][19] Group 4 - Despite the growth potential, there are risks associated with group broadcasting, including regulatory changes and the intense work environment that can lead to health issues among performers [19][20] - The emotional link established through group broadcasting may serve as a new direction for building connections between brands and their audiences, enhancing brand loyalty and engagement [20]

Group 1 - The core viewpoint emphasizes that AI value must be realized through commercialization, as highlighted by the statement from Baidu's CEO, Li Yanhong, indicating that without applications, chips and models cannot deliver value [1] - The AI industry is experiencing a deep differentiation, with major players like Baidu, Alibaba, Tencent, and ByteDance investing heavily to integrate AI into their existing ecosystems, while smaller startups struggle to establish revenue models [1][2] - Major companies are embedding AI capabilities into their products and services, creating a diversified revenue stream and enhancing their existing offerings, as seen with Baidu's Wenxin model and Tencent's integration of AI into its social and office ecosystems [2][3] Group 2 - ByteDance and Kuaishou are finding success in AI commercialization through different strategies, with ByteDance leveraging its product matrix to penetrate various scenarios and Kuaishou enhancing its content ecosystem and commercial efficiency [3][4] - Smaller companies face significant challenges in monetization due to limited resources and market presence, often relying on government contracts or niche markets to survive [5][6] - The commercialization process for startups is slow, with many struggling to convert technology into sustainable revenue, highlighting the importance of finding a balance between technical innovation and market needs [7][9] Group 3 - Establishing a healthy cash flow loop is crucial for both large and small companies in the AI sector, as many face difficulties in user retention and monetization despite a large potential user base [9][10] - The ToB market offers stable customer bases but presents challenges such as high customer education costs and long delivery cycles, making it difficult for startups to compete against established players [10][11] - The focus is shifting from merely having advanced technology to effectively embedding AI into real business applications that generate sustainable cash flow, as seen in the strategies of major companies [12][13] Group 4 - The future of AI commercialization will depend on companies' abilities to integrate their models into business processes and create value, rather than just focusing on technical parameters [13][14] - The remaining players in the AI space will likely be those who can quickly find customers, generate revenue, and adapt to market changes, emphasizing the need for a pragmatic approach to building value [14]

整个科技圈和资本圈都在热议一件事：黄仁勋亲口确认，为中国市场"特供"的H20芯片，在被禁三个月后，恢复销售了！ A股AI概念闻声起舞，英伟达股价应声上涨。一片"利好"声中，硅兔君拨通了自己硅谷专家网络中的专家A（代号无任何意义）的视频电话。A是NVIDIA 前资深系统软件工程师，深度参与过Hopper架构的性能优化项目。 视频那头，A正在硅谷的家中，他呷了一口咖啡，开门见山："Jenson（黄仁勋）的这步棋，走得极其精妙，也极其'毒辣'。它不是一次破冰，而是一场'手 术刀'式的再平衡。 理解了这把刀的落点，你才能看清未来几年中国AI的真实棋局。" 这篇长文，将为您完整转述并解读硅兔君与A长达两小时的深度对话。准备好了吗？让我们一起，拨开迷雾。 H20究竟是"何方神圣"？ A解释说："大模型训练就像成千上万个大脑协同思考一个极其复杂的问题，它们之间需要不断高速交流信息。带宽就是这些大脑之间的'沟通语速'。语速 慢了，整个思考过程就会被拖慢，即便每个大脑本身再聪明，整体效率也上不去。" 这直接决定了H20无法高效地进行万亿参数级别的超大规模集群训 练。 内存带宽（Memory Bandwidth） ：H20配备了 ...

Group 1 - The article highlights the importance of alumni networks in the job market, emphasizing that many companies have preferences for graduates from certain universities due to shared characteristics and specialized skills [2][4][5] - It points out that not all top universities in China maintain strong alumni connections, with many graduates losing touch with their peers shortly after leaving school [2][4] - The article suggests that in a challenging employment environment, leveraging alumni resources can provide unexpected information and opportunities [2][4] Group 2 - The article discusses the differences in alumni activities between Chinese and Western universities, noting that Western alumni events are often more diverse and engaging [5][10] - It mentions that the demand for talent in sectors like e-commerce, new energy, and smart manufacturing is increasing, particularly with the rise of localized operations in overseas markets [10][18] - The article emphasizes the need for professionals to actively explore and innovate in their job search, especially in competitive fields like cross-border business [18][19] Group 3 - The article provides insights from a headhunter's perspective, indicating that companies are increasingly looking for candidates with language skills and cultural understanding as they expand internationally [7][10] - It highlights the importance of understanding local markets and compliance when companies venture abroad, particularly in regions like Southeast Asia [10][18] - The article concludes with a focus on the necessity for job seekers to take control of their career paths and adapt to the evolving job landscape [19][20]

Core Insights - The awarded paper, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," has fundamentally transformed the training of deep neural networks and is recognized as a milestone in AI development [2][3][4]. Group 1: Significance of the Award - The Test of Time Award at ICML honors papers published ten years prior that have had a profound impact on their field, indicating that the research has not only been groundbreaking at the time of publication but has also stood the test of time [3]. - The recognition of Batch Normalization is well-deserved, as it has become a foundational element in deep learning [4]. Group 2: Impact of Batch Normalization - Since its introduction by Google researchers Sergey Ioffe and Christian Szegedy in 2015, the paper has been cited over 60,000 times, making it one of the most referenced deep learning documents of its era [6]. - Batch Normalization has become a default option for developers constructing neural networks, akin to the essential steel framework in building construction, providing stability and depth to models [8]. Group 3: Challenges Before Batch Normalization - Prior to Batch Normalization, training deep neural networks was challenging due to the phenomenon known as "Internal Covariate Shift," where updates in one layer's parameters altered the input distribution of subsequent layers, complicating the training process [12][15]. - Researchers had to carefully set learning rates and initialize weights, which was a complex task, particularly for deep models with saturating nonlinear activation functions [13][15]. Group 4: Mechanism and Benefits of Batch Normalization - Batch Normalization normalizes the inputs of each layer during training by calculating the mean and variance from the current mini-batch, effectively stabilizing the learning process [15][17]. - This method allows for significantly higher learning rates, improving training speed by several times, and reduces sensitivity to weight initialization, thus simplifying the training process [20]. - Additionally, Batch Normalization introduces slight noise from mini-batch statistics, acting as a regularizer and sometimes replacing the need for Dropout to prevent overfitting [20]. Group 5: Theoretical Discussions and Evolution - The success of Batch Normalization has sparked extensive theoretical discussions, with some studies challenging the initial explanation of its effectiveness related to Internal Covariate Shift [21]. - New theories suggest that Batch Normalization smooths the optimization landscape, making it easier for gradient descent algorithms to find optimal solutions [21]. - The concept of normalization has led to the development of various other normalization techniques, such as Layer Normalization and Instance Normalization, which share the core idea of Batch Normalization [25]. Group 6: Lasting Influence - A decade later, Batch Normalization remains the most widely used and foundational normalization technique in deep learning, influencing the design philosophy and thinking paradigms in the field [26][27].

7月初，Meta开出2亿美金"天价"年薪，挖角苹果AI基础模型团队负责人庞若鸣。关于庞若鸣和苹果AI团队的讨论甚嚣尘上：为何Meta愿开高薪挖角一 个"并没有出成果"的团队负责人？ 苹果AI做了两年，成果到底如何？ 《职场Bonus》独家获悉，庞若鸣负责的苹果AI AFM团队一直在做大模型的研发工作，同时也取得了不错的成果，但因苹果内部较为封闭的生态：研发 团队的成果必须要等产品上线后才能发布，成果始终无法公开。 没有人比庞若鸣更惋惜现在的苹果AI。 2021年，庞若鸣加入苹果AI，负责自研模型团队。从2023年Apple Intelligence在内部诞生，到2025年，大模型飞速发展了2年，苹果却没跟上时代的节奏， AI成果频被外界诟病。 7月12日，社交媒体爆出庞若鸣在离职前，向苹果负责软件的SWE部门争取，公开自己AFM团队研发成果的机会，却遭到拒绝。 在迅速变化的AI时代，苹果的AI战略显得保守：2023年以来，WWDC大会上承诺的AI功能频频食言，AI功能虽已融入系统，但却缺乏创新。关于AI的决 策上，苹果已然失去了大众期待的创新和活力。 36氪「职场Bonus」（ID：ZhiChangHongL ...

一位管理培训生向我倾诉了他的困惑："我参与的都是优质项目，"他说，"但项目扎堆反而成了负担。所有事务都找我参与，结果精力过度分散，最终哪 个都做不好。" 我将这种矛盾现象称为"高绩效困境"：企业本想最大化年轻高潜人才的贡献，却在不知不觉中削弱了他们的实际影响力。试想，若这些人才能专注少数核 心任务，本可创造怎样卓越的深度与绩效？然而现实是，组织复杂度持续攀升：冗余流程消耗精力、跨部门协调牵扯资源、模糊制度制造内耗、无效会议 吞噬时间。最终导致生产效率下滑、员工参与度骤降，团队绩效在系统性损耗中不断衰减。 冗长的报告无人问津，周末被琐事缠身，绩效目标高得离谱——企业核心人才正陷入精力透支、士气低迷的困境。如何重燃斗志、提升效能？ 以下七个方法值得借鉴。 全文阅读时长约11分钟 如果企业想保持活力、员工保持投入，变革势在必行。作为波士顿咨询公司（BCG）资深合伙人、世界经济论坛全球领导力学院院长、IMD战略学教授， Knut Haanaes在三十余年咨 询实践与学术研究中，深度追踪了大量"初衷良善却滋生组织内耗"的管理范式。通过以下七个真实案例分享经验：复杂性的根 源往往源自"出于好意"，但若不及时纠偏，反而可能 ...

Core Viewpoint - GAC Group has entered a "wartime state" as it faces significant financial challenges, with expectations of a net loss of 1.82 billion to 2.6 billion CNY in the first half of 2025, marking a shift from profit to loss compared to the previous year [1][2]. Financial Performance - In Q1 2025, GAC reported a loss of 730 million CNY, with projected losses in Q2 expected to range from 1.13 billion to 1.87 billion CNY [1]. - The company anticipates a full-year loss for 2025, which would be its first annual loss since its listing 13 years ago [1][2]. - GAC's net profit for 2024 is projected at 700 million CNY, but after adjustments, the actual loss could reach 2.48 billion CNY [2]. Sales and Market Position - GAC's total sales in June 2025 were 150,100 units, down 8.22% year-on-year, with cumulative sales for the first half of 2025 at 755,300 units, a decline of 12.48% [4]. - GAC's key brands, GAC Trumpchi and GAC Aion, saw significant sales declines of 22.55% and 13.97% respectively in the first half of 2025 [5]. - GAC Toyota is the only brand within the group to show growth, attributed to promotional pricing strategies [7]. Strategic Challenges - GAC's transition to electric vehicles has been slow, with several key models still in the early stages of market penetration, leading to revenue declines due to price wars in the industry [13]. - The company faces structural mismatches in its sales channels, with 70% of sales through traditional 4S dealerships, which are lagging behind in adapting to new sales models [13]. - Internal management issues, including complex decision-making processes and a lack of innovation, have hindered GAC's ability to adapt quickly to market changes [14]. Reform and Future Plans - GAC has initiated a three-year "Panyu Action" plan aimed at improving organizational efficiency and brand collaboration, with a goal of achieving 2 million units in sales by 2027 [17]. - The company is also seeking external partnerships, notably with Huawei, to enhance its product offerings and market competitiveness [20]. - GAC's overseas sales reached 127,000 units in 2024, a 67.6% increase, indicating a strategic push into international markets [20]. Conclusion - GAC Group is at a critical juncture, facing substantial financial losses and market challenges while attempting to implement reforms and adapt to a rapidly changing automotive landscape [23].