Omnilux LED Face Mask Sale: Get A Rare Deal During Prime Day 2025 https://t.co/bXaTu4zwR8 https://t.co/bXaTu4zwR8 via @ForbesVetted ...

If this Wall Street analyst is correct, Nvidia shareholders will see monster returns through the end of the decade.Beth Kindig, lead technology analyst at the I/O Fund, has an impressive track record where chipmaker Nvidia (NVDA) is concerned. In 2021, she correctly predicted the company would surpass Apple's market value within five years. Nvidia checked that box in three years.Earlier this year, Kindig correctly called it a buying opportunity when Nvidia stock crashed after Chinese startup DeepSeek introd ...

Core Viewpoint - Nvidia's market capitalization has surpassed $4 trillion, making it the largest technology company globally, overtaking Microsoft [1][3]. Group 1: Market Performance - Nvidia's stock price rose by 2.8% to $164.42 per share on July 9, marking a significant milestone in its market valuation [1]. - The company became the first in history to reach a $4 trillion market cap, exceeding the total market capitalization of several European countries [3]. - Nvidia's stock has increased over 10 times in value since the beginning of 2023, with a remarkable 89% rise from its April low [3]. Group 2: Growth Drivers - The demand for AI hardware and chips has surged since the launch of ChatGPT in late 2022, contributing to Nvidia's substantial profits [3]. - Analysts predict that annual AI spending will reach nearly $2 trillion by 2028, further driving Nvidia's growth [4]. - Nvidia's GPUs are considered the "gold standard" for AI infrastructure, dominating the data center AI accelerator market [4]. Group 3: Future Outlook - Nvidia plans to build "AI factories" globally, aiming to enhance AI infrastructure and meet the growing demand for AI applications [5][6]. - The company is set to establish the world's first industrial AI cloud facility in Germany, equipped with 10,000 Blackwell GPUs [6][7]. - Analysts have raised Nvidia's target price, with predictions of reaching a market cap of $5 trillion within the next 18 months [4]. Group 4: Policy Environment - Recent changes in U.S. government policy have eased some chip export restrictions, alleviating concerns about Nvidia's business in China [9]. - Despite the positive outlook, some analysts express caution, drawing parallels between the current AI hype and the early 2000s internet bubble [9].

Core Insights - Glaukos Corporation (GKOS) reported a record-breaking first-quarter 2025 performance with revenues of $106.7 million, a 25% year-over-year increase, primarily driven by the adoption of the iDose TR implant [1][8] - The company reaffirmed its full-year revenue guidance of $475-$485 million, indicating confidence in its commercial momentum [1] Near-Term Growth Drivers - The iDose TR launch is gaining traction, with U.S. glaucoma sales increasing by 41% year-over-year, contributing nearly $21 million in the first quarter, reflecting an $85 million annualized run rate [3] - Internationally, Glaukos achieved $29 million in interventional glaucoma sales, a 15% year-over-year increase, supported by the growing acceptance of minimally invasive glaucoma surgery (MIGS) [4] Long-Term Growth Catalysts - Glaukos has a robust R&D pipeline with key programs including Epioxa, iDose Trex, PRESERFLO MicroShunt, and GLK-401, which are expected to broaden the company's addressable market in the next three to five years [5][6] - The company is cultivating standalone use of MIGS devices like iStent infinite, anticipating long-term growth as more surgeons adopt interventional glaucoma practices [7] Challenges and Watchpoints - Reimbursement complexities and LCD restrictions are impacting legacy stent sales, which declined mid-single digits in the first quarter, with expectations of continued pressure through 2025 [8][9] - Sales of Photrexa remain muted due to the Medicaid Drug Rebate Program, with near-term growth in this segment expected to be flat to low-single digits [10] Competitive Landscape - The iDose TR launch positions Glaukos ahead in procedural glaucoma drug delivery, but competition from Alcon, Johnson & Johnson, and Sight Sciences remains significant [11] - Year-to-date, Glaukos shares have declined by 31.7%, underperforming its rivals in the MIGS space, while Alcon and J&J have seen increases [12]

Core Insights - EngageLab participated in The MarTech Summit Hong Kong, a significant event for marketing technology professionals, focusing on trends and innovations in the MarTech industry [1][2] Company Overview - EngageLab is a global leader in customer engagement and marketing technology, providing AI-powered omnichannel solutions that facilitate seamless customer interactions across various channels [8] - The company supports businesses in over 220 countries and regions, delivering more than 1 million messages every second [8] Event Highlights - The summit featured senior marketing executives from renowned brands such as JPMorgan, Yahoo, Nike, and DBS, as well as local enterprises like Cathay Pacific and Hong Kong Disneyland, allowing EngageLab to engage with industry leaders [2] - EngageLab showcased its advanced solutions, including marketing automation and omnichannel communication tools (AppPush, WebPush, Email, SMS, WhatsApp, and OTP), aimed at enhancing customer engagement and driving digital transformation [5][8] Discussions and Opportunities - A senior executive from a leading international airline discussed the challenges and opportunities of implementing an omnichannel engagement strategy with EngageLab, focusing on integrating various communication channels for personalized customer interactions [4] - The airline expressed interest in EngageLab's capabilities in high-concurrency messaging and compliance with data privacy standards for global operations [4] Company Mission - EngageLab aims to empower businesses with scalable and intelligent engagement tools to unlock new opportunities in the digital era, emphasizing innovation and customer-centric growth [6]

智东西 编译 | 程茜 编辑 | 心缘 AI能听懂你的"话外音"了？ 智东西7月8日消息，近日，阿里通义实验室开源多模态推理模型HumanOmniV2。 HumanOmniV2通过引入强制上下文总结机制、大模型驱动的多维度奖励体系，以及基于GRPO的优化训练方法，实现了对多模态信息的全面理解，使得模 型不会错过图像、视频、音频中的隐藏信息，一定程度上规避其在全局上下文理解不足和推理路径简单上的问题。 如在生成最终答案前，模型会输出一个标签内的上下文概括，系统性分析多模态输入内容中的视觉、听觉、语音信号，为后面的推理过程提供依据。如下图 提问"女人为什么翻白眼"，HumanOmniV2基于视频、音频等信息给出正确答案"她的翻白眼更像是对一个潜在敏感话题的夸张、俏皮的反应，非对其他人表 示不满"。 现阶段HumanOmniV2已开源。阿里通义团队还推出包含633个视频和2689个相关问题的评测基准IntentBench，在此之上，HumanOmniV2准确率达到 69.33%。 Hugging Face：https://huggingface.co/PhilipC/HumanOmniV2 IntentBench评 ...

Group 1 - Ruoming Pang, head of Apple's foundational model team, is reported to join Meta's new AI team with an annual compensation in the tens of millions [1] - Pang's departure may be influenced by internal discussions at Apple regarding the introduction of third-party models like OpenAI, leading to team morale issues [1] - Apple's AI team structure will be reorganized under Zhifeng Chen, transitioning to a multi-layer management structure [1] Group 2 - Microsoft has launched Deep Research, a public preview version that utilizes the o3 model and Bing search to create an advanced AI research tool [2] - This AI can automatically deconstruct complex problems, gather the latest authoritative information from the web, and generate auditable research reports [2] - An API interface has been opened for integration into applications, supporting enterprise-level AI platforms across various fields such as research, finance, and healthcare [2] Group 3 - Alibaba has open-sourced the multi-modal reasoning model HumanOmniV2, capable of accurately capturing hidden information in videos and understanding "subtext" [3] - The model incorporates a forced context summarization mechanism, a multi-dimensional reward system driven by large models, and optimization training methods based on GRPO [3] - Alibaba has introduced the IntentBench evaluation benchmark, with HumanOmniV2 achieving an accuracy rate of 69.33%, excelling in understanding complex human intentions [3] Group 4 - PaddleOCR 3.1 has been released, with Wenxin 4.5 enhancing the accuracy of text recognition in 37 languages by over 30%, supporting high-quality automatic data labeling [4] - A new production line, PP-DocTranslation, has been added, combining PP-StructureV3 and Wenxin 4.5 to support translation of Markdown, PDF, and image documents, along with customization of professional terminology [4] Group 5 - A controversy has emerged involving hidden instructions in academic papers aimed at inducing AI to give high scores, with several top universities implicated [6] - Xie Saining, a co-author of one such paper, acknowledged responsibility and apologized, clarifying that he does not endorse such practices [6] - This incident has sparked discussions on academic ethics in the AI era, highlighting the lack of unified standards in AI review processes and the need for reform [6] Group 6 - The Visual Language Action model (VLA) is becoming a core technology for embodied intelligence by 2025, with rapid iterations from Google's RT-2 breakthrough [7] - China's Zhihui Square has partnered with top universities to launch FiS-VLA, innovatively embedding "fast systems" into "slow systems" to address the trade-off between robotic control efficiency and reasoning capability [7] - FiS-VLA has achieved an 8% success rate improvement in simulation tasks and an 11% improvement in real environments, with a control frequency of 21.9Hz, 1.6 times that of the open-source model π0 [7] Group 7 - YouTube co-founder Chen Shijun discussed AI entrepreneurship and long-termism with the Manus team, emphasizing the value of rapid experimentation and risk-taking [8] - Recommendations for AI startups include leveraging first-mover advantages to retain users, creating compound network effects, and exploring areas that larger companies avoid, all within legal boundaries [8] - Key decisions at YouTube included prioritizing user growth over immediate monetization, establishing transparent core metrics, and developing a creator-friendly advertising model while focusing on the "passive experience" of recommendation systems [8] Group 8 - The key shift in acquiring users for AI products is that if a product does not generate social engagement within the first 48 hours, it may fail, making virality a survival threshold rather than a bonus [9] - The success story of selling Base44 for $80 million involved user participation in the development process, encouraging sharing of creations, and strategically choosing LinkedIn as a platform for dissemination, creating a closed loop of development, showcasing, and sharing [9] - The distribution paradigm for AI startups is evolving, with product development becoming a public showcase, niche native creators proving more effective than influencers, and growth metrics becoming assets for dissemination, shifting from "closed-door development" to "public collaboration" [9] Group 9 - U.S. universities are reshaping computer science education, with the CS major potentially becoming more humanities-oriented, emphasizing computational thinking and AI literacy over traditional programming skills [10] - The "Level Up AI" initiative has launched an 18-month curriculum overhaul, where future programming languages may involve "Human," allowing students to complete programming tasks through interaction with AI [10] - Traditional humanities classrooms are facing assessment crises, with educators struggling to identify AI-generated content, leading to a return to handwritten assignments and the development of anti-cheating systems, raising concerns about students' over-reliance on AI affecting their cognitive abilities [10]

HumanOmniV2团队 投稿 量子位 | 公众号 QbitAI 在多模态大语言模型（MLLMs）应用日益多元化的今天，对模型深度理解和分析人类意图的需求愈发迫切。尽管强化学习（RL） 在增强大语言模型（LLMs）的推理能力方面已展现出巨大潜力，但将其有效应用于复杂的多模态数据和格式仍面临诸多挑战。 在深入研究现有技术后，发现在当前多模态推理模型中发现现有的推理路径存在两大核心问题：全局上下文理解不足和捷径问题。 全局上下文理解不足： 当模型无法准确识别或错误解读多模态证据和上下文信息时，便会出现此问题，导致给出不正确的答案。 捷径问题： 指模型在处理多模态输入时，忽视了关键线索，未充分考量多模态信息就直接给出答案，从而导致次优或片面的结果 为彻底解决这些痛点，阿里巴巴通义实验室团队推出 HumanOmniV2 ，强调模型必须在对多模态输入 全局上下文有清晰理解 的 基础上进行推理。这种全局性理解能够有效避免模型遗漏关键多模态线索，确保推理过程的全面性和深入性。 相关代码、模型、数据都开源，地址可在文末获取。 效果展示 问题：这两个人是什么关系？ A. 他们想引起人们对该产品的关注。 B. 这两个人是商业伙 ...

2025年第27周（6月30日-7月6日）科学仪器行业总体发展呈现三个趋势： 技术攻关聚焦国 家战略、 标准与政策双轮驱动、厂商创新注重场景落地 。 以下是一周要闻： 政策与行业规范 · 四川省科研仪器开放共享政策 《四川省重大科研基础设施和大型科研仪器开放共享管理办法》发布。办法中提及建立覆盖全 省的科研仪器共享平台，要求财政资金购置的仪器设备 "应入尽入"，并通过查重机制避免重 复采购。政策明确将考核结果与科技项目申报挂钩，预计提升设备利用率30%以上。 · 市场监管总局推"揭榜挂帅"新模式 市场监管总局发布《食品补充检验方法"揭榜挂帅"工作方案》，通过创新机制激发全社会的研 发力量，增强食品安全检测能力。该方案聚焦于解决当前食品安全监管中的难点，特别是那些 对消费者健康构成严重威胁且检验难度大的问题，如掺杂掺假和非法添加行为，公开征集解决 方案。它鼓励不受限于地域、单位和专业的跨学科团队参与，并承诺对成功的研发成果给予优 先的成果转化支持，确保新方法能迅速应用于实际监管中。 特别提示 微信机制调整，点击顶部"仪器信息网" → 右上方"…" → 设为 ★ 星标，否则很可能无法看到我 们的推送。 · 生态环 ...

7月4日，由深圳上市公司协会"人工智能与智能制造专委会"主办，协创数据（300857）承办的"智能算 力"专题沙龙成功举办。来自深圳行业主管部门、领军企业、投资机构与技术专家以及深上协的会员单 位代表，围绕人工智能与智能制造融合发展的关键议题展开深入探讨。 此次沙龙的成功举办，充分体现出协创数据在人工智能领域的技术实力与行业影响力。协创数据表示 AI技术日新月异，甚至"时新时异"，公司将持续加大研发投入，深化产业链合作，为推动"AI+制造"融 合发展贡献更多创新方案。作为深圳上市公司协会"人工智能与智能制造专委会"副主任委员单位，协创 数据也将持续支持专委会工作，积极参与后续系列活动，与各界携手探索人工智能技术创新与产业应用 的新方向。 值得关注的是，协创数据在算力领域的布局持续加码。日前，公司公告拟使用自有资金向全资子公司协 创云算科技(深圳)有限公司(简称"协创云算")增资9.4亿元，用于扩大算力服务规模、优化数据中心布局 及提升云平台技术能力。此次增资后，协创云算注册资本从1000万元增至9.5亿元，成为协创数据算力 业务的核心载体。截至2025年3月，协创数据已签署不超过30亿元的服务器采购合同，5月 ...