NO.1 电商将成为小红书一级入口 ｜ 2025年8月27日星期三 ｜ 8月26日，36氪报道，今年上半年，盒马NB营业额达到80亿元左右，较去年同期双位数增长。目前总门 店数已近300家，仅半年时间门店数量翻番，主要集中在华东核心城市市区和郊区。盒马方面对该信息 表示不予置评。此前盒马NB推出了"超盒算"自有品牌，其中大米、素食等上百款产品已经更换品牌 名。据悉，近期盒马NB整体也将更名为"超盒算NB "。 8月26日消息，小红书日前启动新版本测试，测试版本中，电商入口位于App主界面。未来，它将成为 小红书App一级入口。新版本显示，App底部导航栏出现了"市集"选项，且紧靠"首页"选项。小红书"市 集"相关负责人表示，"市集"是"小红书生活方式电商"的具象化、产品化呈现。"市集"页面上部，设置 了市集直播、买手橱窗、新品首发等频道。 点评：通过将电商入口提升至一级位置，小红书能够更好地整合内容与电商资源，提升用户购物体验。 此外，市集页面的多样化功能设置，将进一步丰富用户的购物选择。 NO.2 盒马NB上半年营业额达80亿元，目前总门店数近300家 NO.4 豆包上线未成年人模式，默认关闭相关视频展示等 ...

人民财讯8月26日电，记者从字节跳动获悉，豆包App已正式上线未成年人模式。此前，有网友反馈称 因豆包会在问答结果中提供相应短视频作为参考，部分儿童可能借此浏览视频内容，分散注意力。据 悉，在未成年人模式下，回答展示视频、浏览第三方网页、与豆包以外的智能体对话及AI创作等功能 将默认关闭，而翻译、资料查询与深入研究等学习与工作相关能力仍可正常使用。 ...

凤凰网科技讯 8月26日，近日，豆包App已正式上线未成年人模式。据悉，在该模式下，回答展示视 频、浏览第三方网页、与豆包以外的智能体对话及AI创作等功能将默认关闭。而翻译、资料查询与深 入研究等学习与工作相关能力仍可正常使用。 豆包方面表示，将持续优化未成年人模式的体验，帮助大小朋友安心使用AI。 豆包未成年人模式需家长设置密码，如退出或切换仍需二次输入密码确认，避免未成年人绕过设置。 此前，有部分网友反馈称因豆包会在问答结果中提供相应短视频作为参考，部分儿童可能借此浏览视频 内容，分散注意力。经实测发现，未成年人模式开启后，问答中不再有任何短视频相关内容展示或推 荐。此外，用户也无法与豆包主bot外的其他智能体对话。 ...

责任编辑：何俊熹 < 3 未成年人模式 未成年人模式是为家长提供的工具,以方便关闭部分 功能,防止小朋友拿到设备后可能的不适当使用。 A 未成年人模式 0 和豆包以外的智能体对话 回答展示相关视频 C # 浏览第三方网页 E Al 创作 新浪科技讯 8月26日下午消息，近日，豆包App已正式上线未成年人模式。据悉，在该模式下，回答展 示视频、浏览第三方网页、与豆包以外的智能体对话及AI创作等功能将默认关闭。而翻译、资料查询 与深入研究等学习与工作相关能力仍可正常使用。 豆包未成年人模式需家长设置密码，如退出或切换仍需二次输入密码确认，避免未成年人绕过设置。 此前，有部分网友反馈称因豆包会在问答结果中提供相应短视频作为参考，部分儿童可能借此浏览视频 内容，分散注意力。记者实测发现，未成年人模式开启后，问答中不再有任何短视频相关内容展示或推 荐。此外，用户也无法与豆包主bot外的其他智能体对话。 大模型产品作为新兴业态，在青少年数字生活保护上仍有很大探索空间。豆包方面表示，将持续优化未 成年人模式的体验，帮助大小朋友安心使用AI。（罗宁） ...

Core Insights - The project "戒色引流号" operates similarly to the previously analyzed "同城搭子" project, focusing on monetization through content creation and community engagement [1] - The project targets male users seeking self-improvement by presenting themes of positivity and personal growth, establishing a persona of a "spiritual mentor" or "growth partner" [7] Project Overview - The account has published 110 pieces of content and gained over 60,000 followers, indicating effective audience engagement [6] - The monetization strategy involves directing traffic from short video content to private community groups, utilizing a low-cost entry mechanism through QR codes [7][22] Execution Steps - The first step involves obtaining copy from Douyin videos using tools like Doubao App to extract necessary text [11] - The second step is to creatively rephrase the obtained text using AI tools to ensure originality [15] - The third step includes video production using mainstream editing software, applying templates and automated subtitles for efficiency [16] - The final step is to set up an automated group entry system, which can be acquired at a low cost from e-commerce platforms [17] Community Management - Once the system is configured, users can automatically join groups after payment, allowing for a fully automated process [22] - Group maintenance involves regularly sharing relevant materials on the project's theme, enhancing user engagement and retention [22]

Core Insights - The article discusses a project called "Douyin Abstinence Traffic Account," which is similar to a previous concept of "local partners" and aims to generate income through social media engagement [1][6]. Project Logic - The project involves creating positive and growth-oriented content to attract young men seeking self-improvement, establishing a "life coach" persona [7]. - The monetization strategy relies on directing engaged followers into WeChat groups through a paid entry system, facilitated by a QR code on a "traffic card" [7]. Execution Steps - Step 1: Use an app called "Doubao" to extract text from Douyin video links [9]. - Step 2: Modify the extracted text using AI tools to avoid duplication while maintaining the original meaning [11]. - Step 3: Create videos using the "Jianying" app, incorporating a consistent cover design and automatic subtitles [12]. - Step 4: Implement a low-cost tool from Taobao to allow followers to join the WeChat group automatically after payment [13][18]. Additional Notes - Initial performance may not be strong, but consistency in content updates is crucial for eventual success [19].

Group 1 - The research by the Chinese Academy of Sciences confirms that multimodal large language models (MLLMs) can form object concept representation systems similar to humans, providing a theoretical framework for building human-like cognitive structures in AI [1] - The A-share market saw a rebound in AI-related stocks, with significant gains in companies such as Chipone Technology, Tianjun Technology, and Hongsoft Technology, indicating high market interest in AI themes [1] - The Sci-Tech Innovation Board AI ETF (588930) tracks 30 leading AI companies, covering various sectors including computing chips, cloud computing, and robotics, with the top five stocks accounting for 47% of the index weight, suggesting high AI theme purity and elasticity [1] Group 2 - Guotai Junan Securities highlights the significant investment value of the AI sector amid increasing global tech competition, emphasizing the urgency of technological self-reliance in China [2] - The development of domestic EDA tools and advancements in AI research and application capabilities have injected new vitality into the computer industry, with models like DeepSeek R1 nearing international top-tier performance [2] - The upgrade of Doubao App expands AI application scenarios, showcasing the broad potential of AI in daily life, particularly in video chat and Q&A functionalities [2]

Core Insights - The AI sector is experiencing rapid development in evaluation systems, interactive experiences, intelligent applications, and technological innovations, leading to increased competition and accelerated technological iterations, presenting both opportunities and challenges for the future [3]. Group 1: Market Performance - As of June 4, 2025, the Shanghai Stock Exchange Sci-Tech Innovation Board AI Index (950180) rose by 0.16%, with notable increases in constituent stocks such as Cambricon (688256) up 3.53% and Kingsoft Office (688111) up 1.87% [3]. - The Sci-Tech AI ETF (588790) increased by 0.18%, with a latest price of 0.57 yuan and a trading volume of 22.02 million yuan [3]. - Over the past year, the average daily trading volume of the Sci-Tech AI ETF was 271 million yuan, ranking first among comparable funds [3]. Group 2: Fund Performance - The Sci-Tech AI ETF saw a significant growth of 17.7 million yuan in scale over the past week, ranking second among comparable funds [4]. - The fund's shares increased by 6 million over the past week, also ranking second among comparable funds [4]. - In the last four trading days, the fund experienced net inflows on three occasions, totaling 41.94 million yuan, with an average daily net inflow of 10.49 million yuan [4]. Group 3: Financial Metrics - The latest financing buy-in amount for the Sci-Tech AI ETF reached 15.95 million yuan, with a financing balance of 72.43 million yuan [4]. - Since its inception, the fund's highest monthly return was 15.59%, with the longest consecutive months of increase being two, and the longest increase percentage being 26.17% [4]. - As of June 3, 2025, the fund's relative drawdown since inception was 0.40% compared to its benchmark [4]. Group 4: Fee Structure and Tracking Accuracy - The management fee for the Sci-Tech AI ETF is 0.50%, and the custody fee is 0.10%, which are the lowest among comparable funds [5]. - The tracking error over the past three months was 0.015%, indicating high tracking accuracy compared to similar funds [5]. Group 5: Valuation and Top Holdings - The latest price-to-earnings ratio (PE-TTM) for the index tracked by the Sci-Tech AI ETF is 116.72, which is below 82.49% of the time over the past year, indicating a historical low valuation [5]. - As of May 30, 2025, the top ten weighted stocks in the index accounted for 70.6% of the total, with companies like Lanqi Technology (688008) and Cambricon (688256) being significant contributors [5].

Group 1: Market Performance - The A-share market opened high and showed positive momentum, with the Sci-Tech Chip Index performing well [1] - The Sci-Tech Chip ETF (588290) rose by 1.12% in early trading, with a net inflow of 7.31 million yuan over the past three days [1] - Notable stocks included Anlu Technology, which increased by over 4%, along with Huafeng Measurement and Control, Zhongwei Company, and Chip Origin [1] Group 2: Nvidia's Financial Performance - Nvidia reported better-than-expected results for Q1 of fiscal year 2026, with revenue of $44.1 billion, a 69% year-on-year increase, surpassing market expectations of $43.29 billion [1] - Data center revenue for Nvidia reached $39.1 billion, up 73% year-on-year, slightly below the forecast of $39.22 billion [1] - The net profit for the first quarter was $18.78 billion, reflecting a 26% year-on-year growth, with a revenue forecast for Q2 of around $45 billion, with a 2% fluctuation [1] Group 3: Semiconductor Industry Outlook - Tianfeng Securities expressed optimism about global semiconductor growth, particularly driven by AI in downstream applications by 2025 [2] - There is a continuous push for domestic substitution amid escalating risks related to supply chain disruptions and restructuring [2] - The second quarter is expected to show performance elasticity in design sectors such as SOC/ASIC/storage/CIS, with a focus on domestic chip production [2] Group 4: Domestic AI Sector Developments - Guoxin Securities highlighted significant investment value in the domestic AI sector due to rapid advancements and practical applications by local companies [3] - The launch of the Doubao App's real-time video call feature enhances user interaction with AI, showcasing integrated visual and language input capabilities [3] - Kunlun Wanwei's Skywork Super Agents have achieved notable recognition, offering comprehensive content generation capabilities that surpass international competitors in depth and credibility [3]

接近张一鸣的人士说："他会关注 AI 技术细节，自己与顶尖 AI 研究者交流，鼓励字节 AI 人才自由探索。" 文丨贺乾明 王与桐 编辑丨程曼祺 2024 年下半年，市场曾传闻字节跳动正筹建"大模型研究院"。随后，字节回应称："有加强大模型相关 研究的长期计划，但并未决定建立独立机构。" 我们了解到，近期，字节跳动 AGI 研究团队终于浮出水面： 1 月下旬，字节正式设立代号为"Seed Edge"的研究项目，核心目标是做比预训练和大模型迭代更长期、更 基础的 AGI 前沿研究，Seed Edge 已拟定 5 大研究方向。 中国大模型领域目前的一个优势是依托优秀的学习和工程能力，快速跟进行业最新成果：在相对确定的方向 上，用更少资源、更高效率，开发出性能逼近甚至追平领先者的模型。 而 Seed Edge 的成立则是一个不同的信号，标志着字节这家中国互联网科技巨头正试图探索 AI 原创式创新： 用更多资源和人才，摸索尚未确定的智能边界，逼近 AGI 的未来。 张一鸣重视 AI 研究投入，Seed Edge 设立 5 大研究方向 AI 是字节目前最看重的技术和业务方向。 接近字节的人士告诉我们，字节创始人张一鸣非 ...