3 6 Ke
Search documents
对联用 AI 写,测完我真是醉了...
3 6 Ke· 2026-02-21 01:42
Core Viewpoint - The article discusses a creative competition among various AI models to generate couplets for the upcoming Year of the Horse, targeting different demographics: elderly, young adults, and elementary school students. The performance of each model is evaluated based on their ability to meet the specific needs and preferences of these groups. Group 1: Elderly Audience - The elderly prefer couplets that convey stability and traditional values, focusing on themes like health and family happiness [5][12]. - The AI model "Doubao" produced a well-rounded couplet that met traditional expectations, while "Deepseek" impressed with its cultural depth [6][8]. - "Gemini 3" provided a straightforward yet relatable couplet, while "ChatGPT" was criticized for being overly conventional and lacking creativity [11][12]. Group 2: Young Adult Audience - Young adults seek couplets that are trendy, relatable, and motivational, incorporating modern language and internet culture [13][14]. - "Doubao" effectively captured the aspirations of young workers with its use of popular phrases, while "Gemini 3" was praised for its clever integration of relatable themes [17][21]. - "Deepseek" was noted for its visual appeal but did not resonate as strongly, and "ChatGPT" was seen as too rigid and lacking in contemporary relevance [24][25]. Group 3: Elementary School Audience - For elementary students, the couplets should be simple, fun, and easy to understand, focusing on themes of learning and growth [25][26]. - "Doubao" created a clear and engaging couplet suitable for both children and parents, while "Deepseek" produced a playful couplet that resembled a children's song [27][28]. - "Gemini" also performed well, creating a lively image of a smart and eager child, whereas "ChatGPT" was deemed too formal and lacking in charm [30][31]. Group 4: AI Models' Self-Evaluation - The AI models engaged in a self-assessment round, with "Deepseek" critiquing others for their strengths and weaknesses, particularly highlighting "ChatGPT" for its lack of modern appeal [32][34]. - "ChatGPT" countered by pointing out significant errors in "Deepseek's" output, emphasizing the importance of accuracy in couplet creation [36][39]. - The final rankings indicated that "Doubao" was the most balanced, "ChatGPT" was traditional but safe, "Gemini 3" was creative but less formal, and "Deepseek" suffered from critical errors [40][44].
北上广“牛马”,羡慕县城同学
3 6 Ke· 2026-02-21 01:05
春节,像一个巨大的坐标系,让散落在天南海北的年轻人们,短暂回归原点。当一线城市的打拼者与留守县城的老同学在饭桌上重逢,一场无声的比较便 开始了。 总有一些瞬间,会让在一线城市打拼的年轻人,心中涌起对老家同学的羡慕:或许是发现老家同学的月供比自己在一线城市的房租还便宜,或许是看到他 们晚上六点准时下班能享受自己的时间,而自己的手机相册里,除了公司就是地铁站。 这种羡慕,起初是源于物质上的对比——宽敞的房子、无忧的消费。但仔细一想,是一种"确定感"。在小城,人生的轨迹更清晰:稳定的工作、安稳的家 庭、可预期的未来。而在大城市,每一个选择都伴随着一定风险。 但有意思的是,这份羡慕似乎是相互的。县城的同学也会打探互联网大厂的八卦,羡慕他们朋友圈里的都市见闻,羡慕他们的见识和更多的可能性。 我们似乎都活在对彼此生活的想象里,不自觉地美化着自己没走过的那条路。大城市有它的机遇和代价,小城市有它的安稳和束缚。 这种偶尔的"羡慕",可能只是我们在疲惫时一个短暂的情绪出口,也可能是认真考虑的"备选方案"。我们在一次次比较中,更清楚自己想要什么、不想要 什么,也最终学会接纳自己的选择,以及这一路上得到的与失去的。 1 十年北漂: ...
警惕初创公司“Palantir化”:如果没有核心平台,你只是一个穿了西装的埃森哲
3 6 Ke· 2026-02-21 00:03
Core Insights - The article discusses the trend of startups adopting a "Palantir-like" model, emphasizing the deployment of Forward Deployment Engineers (FDEs) to create customized workflows for clients, akin to special forces rather than traditional software companies [1][2][10]. Group 1: The Appeal of the Palantir Model - The allure of the Palantir model lies in its ability to integrate disparate systems and deliver customized platforms quickly, which is particularly appealing to startups aiming for significant contracts [2][10]. - The surge in FDE job postings, with reported increases of 800% to 1000% this year, indicates a growing trend among AI startups to ensure effective deployment through on-site engineering support [10][12]. Group 2: Unique Attributes of Palantir - Palantir's success is attributed to its unique combination of a robust product platform, embedded elite engineers, and proven performance in critical government and defense environments, which most companies struggle to replicate [9][21]. - The company's approach is not merely about automating existing processes but involves pushing clients to adopt new workflows, which is a rare courage among vendors [18]. Group 3: Challenges of Imitating the Model - Many startups attempting to emulate Palantir may end up as high-cost service firms without the competitive advantages that come from a true platform [2][10]. - The article warns that not every problem warrants a "Palantir-level" solution, especially in less critical environments where the ROI does not justify extensive customization [15][16]. Group 4: Key Considerations for Adoption - Companies should assess the criticality of the problems they are solving, the concentration of their customer base, and the regulatory environment before adopting a Palantir-like model [23]. - The article suggests that startups should view FDE deployment as scaffolding rather than the foundation of their business, ensuring that they do not become trapped in a service model [25][30]. Group 5: Practical Steps for Implementation - Startups are encouraged to focus on building strong foundational components rather than fully customized workflows, allowing for scalability and reusability [26]. - Engaging FDEs as integral parts of product development rather than just delivery can create a feedback loop that enhances product quality and customer satisfaction [27].
王腾回应换iPhone17,称因适配睡眠监测设备
3 6 Ke· 2026-02-20 15:35
Core Viewpoint - Wang Teng, the founder of Today Sleep Technology, has publicly switched to using an iPhone 17 for better compatibility with sleep monitoring devices, marking his first use of a non-Xiaomi phone since leaving the company in September 2025 [1] Group 1: Company Transition - Wang Teng previously served as the General Manager of Xiaomi's China Marketing Department and General Manager of the REDMI brand before founding Today Sleep Technology [1] - The company, Today Sleep Technology, was established in January 2026 and has completed seed round financing of several million yuan [1] Group 2: Product and Market Insights - Wang Teng highlighted that sleep is a complex system requiring multifaceted improvements, sharing his sleep data from the past six months [1] - The switch to iPhone 17 was specifically due to better adaptation with sleep monitoring devices such as Whoop, Oura, and Eight Sleep [1]
春晚“带货”新物种:具身机器人何以被疯抢
3 6 Ke· 2026-02-20 09:03
这不是简单的"春晚带货",这是具身智能产业的一个历史性时刻——2026年,这个行业已走出概念验证的"前夜",正在加速商业化落地。 从"一枝独秀"到"四强争霸",行业迎来"GPT时刻" 在四大机器人企业"霸屏春晚"之前,业内便早有共识:2026年,中国具身智能产业将继续爆发增长——此前,摩根士丹利报告就指出,2026年将成为中国 市场爆发的关键拐点;IDC预测,中国人形机器人应用场景将升至当前的3倍以上;瑞银则预计,全球人形机器人需求量在今年将达到3万台。 春晚的火爆,不过是行业爆发的"催化剂"——当国民级舞台成为技术试金石,机器人产业的商业化进程,被按下了"加速键"。 本文约4100字,建议阅读6分钟 当春晚的掌声渐渐散去,一个更真实的商业故事悄然上演。 除夕当晚,魔法原子的100只春晚限定"机器熊猫"上线即售罄,001号更在两小时内以5.7万元的价格被北京网友拍得,其全线产品——从MagicDog到人形 机器人Z1开发版——全部显示"无货"。 同一时间窗口,京东平台机器人搜索量环比增长超300%,订单量增长150%,新增订单覆盖全国100多座城市,从一线到县域,从极客玩家到普通消费 者。他们用真金白银投出的 ...
4天破17亿的《飞驰人生3》,拿下春节最爽电影
3 6 Ke· 2026-02-20 07:42
基本已经毫无悬念了。 《飞驰人生3》轰鸣的引擎,正创下2026马年春节档的最强音。 上映4天拿下17亿票房,韩寒导演的《飞驰人生3》基本已锁定今年春节档冠军,相当于开战便锁定胜局。 | · · | | | | | --- | --- | --- | --- | | 全国总票房 | | | | | 50.00 | | | | | 数据截至2026-02-19 11:14:06 总出票1.15亿张 总场次2082.6万场 | | 平均票价43.2元 | | | (00) 票房排名 | | | | | 影片 年票房 票房占比 排片占比 (万) | | | 上座率 | | 飞驰人生3 124420.53 2026-02-17上映 | 24.8% | 4.4% | 16.8% | | 疯狂动物城2 54786.64 2025-11-26上映 | 10.9% | 15.1% | 3.5% | | 匿杀 43583.05 2025-12-31上映 | 8.7% | 12.9% | 3.6% | | 阿凡达3 42182.54 2025-12-19上映 | 8.4% | 7.3% | 3.6% | | 惊蛰无声 42147. ...
春节假期玩什么:赛博登山潜水、看太空歌剧,还有派对、AVG和JRPG
3 6 Ke· 2026-02-20 07:31
春节假期,来自触乐编辑们的游戏推荐又来了!我们会各自推荐一款假期近期上线的新游戏(也可能是经典老游戏以及值得关注的冷门游戏),它们可能 是PC或主机游戏,也可能是移动游戏网页游戏,来供大家参考和选择。 俊美的少男少女 《永恒终焉》最有特色的部分是战斗系统。游戏中有3个可操作的主角,每个主角都有独立的血条,但大多数情况下血量见底并不会导致角色直接死亡, 而是会消耗屏幕下方的I.S.宝石"续命",所以只要还有I.S.宝石,3位主角就能持续活动下去;游戏战斗过程采用半即时制,玩家不操作角色行动时,敌人 也不会动,你可以仔细观察战局形势;战斗时,敌我双方都使用枪来攻击,但角色仅靠站桩射击无法打败敌人,必须灵活使用核心输出机制"无敌攻 击"——无敌攻击需要消耗一颗I.S.宝石发动,使用过程中,角色不会受到伤害。 侯雯钊:《永恒终焉》(多平台) 关键词:日式、枪斗术、半即时制 一句话推荐:JRPG的一朵奇葩。 《永恒终焉》(End of Eternity,简称EOE)是一款由tri—Ace开发、Sega发行的JRPG,最初于2010年发售,2018年推出了4K高清版本,现在各平台上能 买到的应该都是高清版。另外,游戏只在 ...
李飞飞,刚刚又融70亿
3 6 Ke· 2026-02-20 07:31
Core Insights - World Labs, founded by Fei-Fei Li, has completed a new financing round of $1 billion, bringing its valuation to $5 billion within just 16 months of establishment [1][2][4]. - The company has attracted significant investments from major players like NVIDIA and AMD, as well as new investors including Autodesk, which invested $200 million [3][4]. - World Labs aims to focus on robotics and scientific discovery, actively recruiting engineers to develop core technologies for autonomous navigation and 3D reconstruction [4][5]. Company Overview - World Labs was founded by Fei-Fei Li, a prominent figure in AI, who previously initiated the ImageNet project that significantly advanced deep learning in computer vision [5][6]. - The company has raised a total of approximately $2.3 billion in funding since its inception, with its latest round pushing its valuation to $5 billion [4][6]. - The founding team includes notable scientists and alumni from prestigious institutions, reflecting a strong academic and professional background [6][8]. Product and Technology - World Labs launched its first commercial product, Marble, which generates persistent 3D worlds from various input formats using advanced technologies like Neural Radiance Fields (NeRF) [4][5]. - The business model is based on a freemium approach, allowing users to generate content for free up to four times, with paid plans available for $95 per month [4]. - The company is set to collaborate with Autodesk to enhance its research and modeling capabilities, aiming to build physical AI that enhances human creativity [5][8]. Industry Context - The AI sector is experiencing a wealth creation surge, with companies like Anthropic and Databricks achieving valuations of $380 billion and over $134 billion, respectively [9][11]. - In China, AI companies like Zhiyuan and MiniMax have seen their market capitalizations soar, reflecting the global trend of rising valuations in the AI industry [11]. - The current investment landscape prioritizes "AI content" as a key evaluation criterion for projects, indicating a strong market focus on AI-driven innovations [11].
谷歌Gemini 3.1 Pro新王登场,一口气手搓Win11操作系统,造出模拟城市app,SVG效果绝了
3 6 Ke· 2026-02-20 05:33
Core Insights - Google has officially launched its new flagship model, Gemini 3.1 Pro, which outperforms previous models in 12 benchmark tests, including Gemini 3 Pro, Claude Opus 4.6, and GPT-5.2, achieving the top position [1][29]. Benchmark Performance - In the benchmark tests, Gemini 3.1 Pro scored 44.4% in the "Humanity's Last Exam" academic reasoning test, surpassing Gemini 3 Pro's 37.5% [2]. - The model achieved a remarkable 77.1% in the ARC-AGI-2 abstract reasoning puzzles, doubling the performance of Gemini 3 Pro [2][29]. - For scientific knowledge, Gemini 3.1 Pro scored 94.3% in the GPQA Diamond test, indicating strong performance in factual accuracy [2]. Model Enhancements - The enhancements in Gemini 3.1 Pro focus on complex task processing capabilities, including advanced reasoning, multimodal understanding, and project generation [10][29]. - The model's ability to generate detailed SVG animations and interactive designs has significantly improved, showcasing its capability in creative programming and interactive design [21][23]. User Access and Pricing - Starting today, Google AI Pro and Ultra subscribers can access Gemini 3.1 Pro through various platforms, while free users can ask two questions [10]. - The API pricing for Gemini 3.1 Pro follows a tiered model, with input prices set at $2.00 per million tokens for prompts under 200,000 tokens, and $4.00 for those exceeding this limit [10][11]. Industry Trends - The AI industry is shifting from general capability comparisons to real-world complex task execution, with companies focusing on enhancing reasoning, engineering, and multimodal understanding [33]. - Google's recent rapid advancements, including the launch of Gemini 3 Deep Think and Gemini 3.1 Pro, emphasize the importance of developing models that can address complex real-world problems [33].
一年流入9800名百万富翁:迪拜凭什么?
3 6 Ke· 2026-02-20 05:27
"就这几年,光中国客户都忙不过来……" 前几天,我和一个在迪拜打拼的聊天时,他这样感慨。 六七年前他还在北京做工商财税服务,后来机缘巧合下,竟然跑到中东阿联酋的迪拜,做起了企业服务、信息咨询和房产中介。 他的日程每天都能排得满满当当。 "新来的客户太多了,尤其是科技和金融领域的。他们出手很快,看中一套办公室或者公寓,几乎不怎么犹豫……" "从疫情前到现在,迪拜整体房价涨了差不多70%。"他说,"这还只是一个平均数,一些热门地段的豪宅,涨幅早就翻倍了。" 世界真有点疯狂。 "疯狂"的切面 迪拜的房价,这几年是全球顶级表现。 有数据为证。 根据全球房地产指南(Global Property Guide)的数据,迪拜的房价从疫情前低谷以来,确实走出了一条陡峭的上扬曲线。 仅2023年,迪拜的住宅交易额就飙升了56%,2024年更是打破以往纪录,交易量达到18.1万笔,同比增长36.5%。 迪拜公寓与别墅销售量和中位数价格 REIDIN发布的报告显示,截至2025年12月,迪拜住宅市场销售价格指数同比增长了12.88%,其中别墅(+15.16%)还要比公寓(+12.52%)表 现得更亮眼。 房价只是个切面,它的底层逻 ...