[{"data":1,"prerenderedAt":667},["ShallowReactive",2],{"content-\u002Fposts\u002Fsurvival-welcome-to-university":3,"blog-posts":197},{"id":4,"title":5,"body":6,"categories":182,"date":183,"description":184,"draft":185,"extension":186,"meta":187,"navigation":188,"path":189,"seo":190,"stem":191,"tags":192,"__hash__":196},"posts\u002Fposts\u002Fsurvival-welcome-to-university.md","生存手册——欢迎来到大学",{"type":7,"value":8,"toc":172},"minimark",[9,13,16,19,24,27,30,33,36,42,45,51,54,57,60,63,69,72,75,78,81,84,87,90,93,96,103,106,109,112,115,118,121,124,127,130,136,139,142,145,151,154,157,160,163,166,169],[10,11,12],"p",{},"2008 年，几个上海交大本科生写了一本 62 页的生存手册，开篇第一句是：\"国内绝大部分大学的本科教学，不是濒临崩溃，而是早已崩溃。\"",[10,14,15],{},"它真正耐用的判断在更靠后的地方：\"在你毕业之时，那一串苍白的分数其实就已经作废了。\"这句判断在 2008 年就已经写出来了，比大多数关于大学的经济学讨论更早问到了同一件事：大学在教什么，大学在筛什么。它离一个正在读书的人最近的时刻，是排保研名次、投简历、被用人单位挑的时候。那本手册写给 2008 年刚入学的人，这句话今天照样适用。",[10,17,18],{},"这两个问句指向的东西不同。教什么，看的是四年之后你身上留下了什么；筛什么，看的是雇主凭什么在一群人里把你挑出来。这两件事在理想情况下是同一件事，现实里常常不是。",[20,21,23],"h2",{"id":22},"不拿学位但获得教育你选哪个","不拿学位但获得教育，你选哪个",[10,25,26],{},"先看一个说不通的选择：一边是不拿学位但获得教育，一边是得到学位却无法接受教育。多数人会选择后者。",[10,28,29],{},"如果上学的主要目的是学习，这个选择说不通。学位只是对所学内容的一个粗略估量，但学生为了拿到它，会策略性地选\"轻松得 A\"的课，雇主则几乎不关心这些课具体是什么。一边是学生按给分松紧挑课，一边是雇主不看课程内容，两边对着同一个东西出手，而那个东西不是课上教的内容。",[10,31,32],{},"再看一组数据。全球范围内，个人多受一年教育，未来收入预计上涨 8% 到 12%；而国家层面，公民平均多受一年教育，国民收入仅增长 1% 到 3%。",[10,34,35],{},"如果教育主要在提升生产力，个体多读一年，宏观上就该一起涨。事实没有。它能说明的只有一件事：个人收益里有相当一部分来自他在人群里的位置，而不是他生产出来的东西。",[37,38,39],"note",{},[10,40,41],{},"这两个数来自不同口径：8% 到 12% 是按个体工资算出来的私人教育回报率，1% 到 3% 是把国民平均受教育年限放进跨国增长回归得到的系数。搭在一起不能算出教育对国家的回报率。宏观这一侧的估计会被教育数据的测量误差压低，Krueger 和 Lindahl 在 2001 年综述教育与增长的文献时，结论反而是宏观系数至少不弱于微观回报率。把这两组数字并置起来的论证方式，更接近 Bryan Caplan 在《The Case Against Education》里的用法。",[10,43,44],{},"Hungerford 和 Solon 在 1987 年发现了一个更直接的证据：完成四年学位的人与读了三年零十一个月但未获学位的人之间，存在显著的薪资差距，这个差距无法用多出来的那一个月教育来解释。这件事不排除能力差别和雇主的偏好，它只是把\"证书\"这件东西单独拎出来，交给下一节去解释。",[46,47,48],"tip",{},[10,49,50],{},"这篇论文发表在 1987 年的《Review of Economics and Statistics》，标题里用的词是 sheepskin effects，字面意思是羊皮纸，也就是文凭本身。用这一现象解释学历价值的文献通常归到信号假说那一侧，它最初要解释的正是\"多读的那一年拿到证书\"和\"多读的那一年没拿到证书\"之间的收入差。",[20,52,53],{"id":53},"学历的信号价值随持有人数量下降",[10,55,56],{},"多出来的那一个月解释不了这份差距，差额落在那张纸上。这么看，雇主买的是别的东西。",[10,58,59],{},"经济学家罗宾·汉森和凯文·西姆勒在《脑中的大象》第 13 章给出了一个解释框架，起点是一套标准的人力资本叙事：学生上学是为了提升自我，雇主愿意为提升过自我的员工付高薪。",[10,61,62],{},"这套说法解释不了上一节的三个现象。1973 年，Spence 提出的信号传递模型提供了一条替代解释。",[37,64,66],{"title":65},"信号传递模型的两个前提",[10,67,68],{},"模型能成立靠两件事：雇主相信学历和更高的工作能力相关，而能力低的人拿到学历的成本更高。第二个前提是信号可信的关键，缺了它，学历就只是一张谁都拿得到的纸。原模型只有一个未观测变量，就是能力；智力、责任心、能否延迟满足这些是后来的经验研究往里填的候选特质，模型本身对教育有没有提升生产力并不表态。",[10,70,71],{},"在这个模型里，雇主无法直接观测求职者的智力、责任心、延迟满足的能力和遵守规则的习惯，而这些特质对工作表现很重要。学历恰好提供了一个可用的代理指标：能够完成多年学业、通过一系列考试、遵守学校纪律的人，大概率具备这些特质。课上读到的东西没有被雇主看到，被看到的是这个人能不能被训练；而这个判断能成立多久，取决于学位持有人有多少。",[10,73,74],{},"于是学生读四年书，真正在做的事情就是向未来的雇主发送一组可比较的信号。雇主用学位完成记录、学校层级和成绩排名来筛选，不是因为它们与工作内容直接相关，而是因为它们能低成本地预测一个人是否可培训。",[10,76,77],{},"信号之外，还有一个更少被讨论的功能：驯化。信号能解释雇主为什么看学历，解释不了教学质量不升而认证照旧这回事，因为学校在这里还有第二项产出。书中转述了一项观察，欠发达地区的未受教育工人在简单重复劳动中效率低下，不按时可靠地工作，偏好间接暗示而非直接命令，难以接受与其预设文化身份冲突的任务；学校在发达社会中系统性地矫正了这些行为。",[10,79,80],{},"老师奖励遵守纪律，训练学生接受测量、评分和排名，十多年的训练把\"按时出勤、完成截止任务、服从命令链\"变成默认习惯。",[10,82,83],{},"这套习惯进入雇佣关系后，雇主获得的是一个已经被预训练过的劳动力。信号解释雇主为什么挑学历，驯化解释学校为什么能持续供给被雇主接受的劳动力，两者合起来，教育扩张就可以在教学质量不升的情况下照旧进行。",[20,85,86],{"id":86},"几家公司开始不用这张证书",[10,88,89],{},"前两节讲的是一套还在运转的机制。2025 年开始，招聘端有一部分公司选择绕开它。",[10,91,92],{},"2025 年，深圳 17 岁高中生陈广宇因持续在技术社区进行高质量技术分享与写作，以实习生身份加入 Kimi 团队。他在\"注意力残差\"研究中被标注为\"共同第一作者\"，这项研究正是支撑 Kimi K3 的两项关键架构创新之一。Kimi K3 以 1679 分登顶 Arena.ai 的 Frontend Code Arena 前端编程榜单，马斯克在这条榜单下留言\"Impressive\"。",[10,94,95],{},"这不是孤例。Kimi 背后的月之暗面随后推出\"穿越计划\"，针对 2027 届毕业生，首批开放 16 个名额，入选者通过 3 到 6 个月实习考察后，即使尚未毕业也直接获得公司期权，不限专业、不限学历、不限经验。Palantir 在 2025 年开启高中生招聘计划，直接进入核心部门做主力产品研发，500 人报名只录取 22 人。",[97,98,100],"caution",{"title":99},"三组被传得比原始报道更满的说法",[10,101,102],{},"「穿越计划」的一手报道来自《晚点》，官方口径是拟推出、拟授予，3 到 6 个月考察期只在二手摘要里出现。Palantir 那条可核验的是第二期每月 5,400 美元津贴，转正年薪 17 万美元只见于中文二手报道。Google、Meta、OpenAI「两年累计录用 300 多名未完成大学学业的年轻人」找不到原始出处，也没有说明\"录用\"是实习还是全职，本节没有采用这一条。前两条按原稿保留，第三条已在正文中删去。",[10,104,105],{},"这两家不再把学历当作判断人的入口，改用能直接看到的产出和测试结果。它们挑人的依据变了，招聘的门也就跟着开了。",[10,107,108],{},"陈广宇被 Kimi 录用，不是因为他的学历，是因为他在技术社区持续输出的技术内容。Palantir 招的高中生需要有编程和统计基础，还要通过 4 个月的专项培训和实战项目考核才能转正。两家看的是候选人实际做过什么、能做什么，而不是他读了多少年书。",[10,110,111],{},"李飞飞公开表示，招聘软件工程师时更看重候选人的 AI 技能而非大学学位。DeepSeek 的 Harness 团队负责人崔添翼在微博上回应招聘问题时表示，岗位描述中未说明则不存在硬性学历门槛，刚完成高考的优秀高中生也可以申请实习。",[10,113,114],{},"按技能招人已经不是个别公司的偏好，而是招聘口径在改。腾讯副总裁蒋杰观察到，在校年轻人做的方案与工业界的代差越来越小，对大模型的理解已经和工业界非常接近。",[10,116,117],{},"当高校培养体系从领先产业半步变成滞后产业一步，企业等待毕业生\"成熟\"的成本变得不可接受。缺口不大的时候，课程更新慢一两拍并不致命；AI 行业把这个缺口拉大到以年计，企业就开始自己从高中生和在校生里挑人。",[20,119,120],{"id":120},"信号失灵是有条件的",[10,122,123],{},"学历作为信号之所以能起作用，前提是它能把人区分开。录取率很低、学位持有人稀少的时候，一个本科学位就能告诉雇主：这个人大概率具备基本智力、责任心和服从能力。",[10,125,126],{},"高等教育扩张把学位持有人的数量抬上去，同一个本科学位能分出的差距就越来越小。录取率从百分之几涨到接近普及，硕士又接着扩招，雇主手里那张筛子的孔就越换越细。当几乎所有人都有本科学位，它就不再能区分任何人，而 AI 行业是第一批撞上这件事的行业：那里的技能水平差异可以当场测出来，学历这道代用的筛子显得多余。",[10,128,129],{},"世界经济论坛《2025 年未来就业报告》给出一个宏观口径：81% 的雇主在招聘中评估技能时预计优先考量工作经验，43% 预计会依赖大学学位的完成情况。Indeed 的数据显示，2024 年 1 月五成二的招聘岗位已无学历要求，要求本科的岗位五年间从两成下降至一成八。这两组数字支持的是雇主对学位的依赖在下降，并且都只是雇主自报的招聘口径，不等于学位已经不起作用。",[131,132,133],"important",{},[10,134,135],{},"这两个数字回答的是\"评估技能时会优先采用哪些方式\"，是雇主对 2025 到 2030 年的自报偏好，不是招聘的实际结果。同一份报告里还有一条相反的观察：15 个增长最快的岗位里，14 个仍然主要要求大学学位。Indeed 那一组的单位是招聘广告文本，看的是广告里有没有提到正式学历要求。",[20,137,138],{"id":138},"那么你该怎么办",[10,140,141],{},"这对学生的行为指导不是\"不要上大学\"，而是理解大学在做什么，然后决定自己需要什么。",[10,143,144],{},"大学同时生产三样东西：可验证技能、认证信号、驯化资本。可验证技能就是前面说的\"教什么\"，对应那个不拿学位但获得教育的选择；认证信号是雇主用来筛人的那张纸；驯化资本是这套训练留给你的工作习惯。比例因专业、学校和目标市场而异。",[46,146,148],{"title":147},"怎么估自己那一样占比更高",[10,149,150],{},"看你要进的那个市场，用它验证能力的成本来判断。验证成本低的地方，也就是代码能跑、证能考、手术台上有带教的岗位，雇主会直接测，技能那一项的权重就高。验证成本高、或者一次招错代价大的入口岗，雇主会退回学历和实习经历这类代用指标，信号那一项更值钱。驯化资本在所有雇佣关系里都算分，只有在需要你自己定方向、而不是执行既定流程的岗位上，它才从资产变成负债。",[10,152,153],{},"如果目标市场有硬技能验证机制，例如软件开发的代码测试、会计的执业考试、医学的执照和住院训练，技能价值占比更高，纯粹依赖学历证书的策略回报递减。",[10,155,156],{},"如果目标市场是依赖学历筛选的传统白领入口岗位，学历信号仍有价值，但竞争者在增加，信号的区分度在下降。",[10,158,159],{},"驯化资本在几乎所有雇佣关系中都有价值，但过度驯化在需要独立判断和创造性输出的岗位上可能成为负债。",[10,161,162],{},"三条路合起来是一句：学位照拿，但不要把全部筹码押在它上面。读书期间持续做出别人能看到的产出，把可验证的东西攒在手上，这件事从今天就能开始做。",[10,164,165],{},"陈广宇的经历提供了一个具体的参照。他在中学时兴趣广泛，借用家长身份注册 Shopify 店铺，组织青少年开源社区，做短视频账号四天涨了 29 万粉丝。这些经历本身不走同一条路，共同点是每一件都在真实场景里留下过别人能看到的产出，他后来被看见靠的是这个。",[10,167,168],{},"他在一篇自传性文章里写道：\"我先前的所有经历并没有浪费，它们训练了我一个能力：怎么快速进入一个陌生领域，并找到前进的路径。\"他靠的不是提前学完大学课程，而是持续在真实场景里做事。",[10,170,171],{},"2008 年那个\"分数作废\"的判断，说的其实就是这件事：分数作为信号载体，在发出之后对个体就不再具有私人价值；学位作为筛选装置，在雇主端的流通效率也在下降。两者可以同时成立。",{"title":173,"searchDepth":174,"depth":175,"links":176},"",2,3,[177,178,179,180,181],{"id":22,"depth":174,"text":23},{"id":53,"depth":174,"text":53},{"id":86,"depth":174,"text":86},{"id":120,"depth":174,"text":120},{"id":138,"depth":174,"text":138},"生活杂谈","2026-10-12 20:00:00","大学同时生产三样东西：可验证技能、认证信号和驯化资本。你付了四年的钱，真正拿到手的是哪一样，取决于你毕业时要进的那个市场在为什么付费。",false,"md",{},true,"\u002Fposts\u002Fsurvival-welcome-to-university",{"title":5,"description":184},"posts\u002Fsurvival-welcome-to-university",[193,194,195],"大学","求职","方法论","G0mR9-sJt_77WoFwsq6MWuT5H50D8WdZVSA66gew8oQ",[198,200,205,209,218,224,228,238,248,254,260,269,273,277,281,291,298,304,308,313,321,328,336,341,347,353,358,363,370,377,384,390,397,405,411,417,423,429,434,442,449,457,463,469,476,483,490,497,504,510,516,523,529,536,544,552,558,565,572,579,585,593,599,605,612,619,625,633,638,645,652,659],{"path":189,"title":5,"description":184,"date":183,"categories":182,"tags":199,"draft":185},[193,194,195],{"path":201,"title":202,"description":173,"date":203,"categories":173,"tags":204,"draft":185},"\u002Fposts\u002Fgp-tab-测试","gp-tab-测试","2026-10-12 10:51:22",null,{"path":206,"title":207,"description":173,"date":208,"categories":173,"tags":204,"draft":188},"\u002Fposts\u002Fnetwork-clean","怎么让网络变得干净","2026-10-11 19:40:02",{"path":210,"title":211,"description":212,"date":213,"categories":182,"tags":214,"draft":185},"\u002Fposts\u002Fthe-age-of-hesitation","[搬运] 不敢投入的时代","为什么我们现在越来越不敢投入了。很多时候我们以为的选择，比如你有两个工作offer，现在选择好像是我到底应该考公还是考编。明明这两个选项都不是真正的选项，因为你还没有开始，也不知道最后能不能成。这个时候我们就已经开始犹豫了。","2026-10-10 11:44:12",[215,216,217],"翻译","搬运","心理学",{"path":219,"title":220,"description":221,"date":222,"categories":182,"tags":223,"draft":185},"\u002Fposts\u002Fwhy-happiness-look-like-same","[搬运] 为什么大家的幸福都长一个样？","为什么大家的幸福都长一个样？一种认知被广泛传播后似乎变得不断扁平，这种感觉很奇怪。","2026-10-09 13:57:15",[215,216,217],{"path":225,"title":226,"description":173,"date":227,"categories":182,"tags":204,"draft":185},"\u002Fposts\u002Feverything-is-paid-twice","任何事情都要被消费两次","2026-10-06 12:30:06",{"path":229,"title":230,"description":231,"date":232,"categories":233,"tags":234,"draft":185},"\u002Fposts\u002Fwhy-llm-seems-worse-after-launch","为什么模型上线后，体感像是“降智”了？","权重与 KV cache 的量化都能降低 serving 成本，也可能改变输出；但体感变弱不足以证明服务后来切换了精度。","2026-09-29 21:44:08","技术杂谈",[235,236,237],"LLMs","Serving","量化",{"path":239,"title":240,"description":241,"date":242,"categories":243,"tags":244,"draft":185},"\u002Fposts\u002Fspinning-up-policy-optimization","[译] 深度强化学习入门（三）：策略优化入门","OpenAI Spinning Up 教程第三部分的全文翻译：推导最简单的策略梯度、log-derivative trick、EGLP 引理、reward-to-go 与基线，配 PyTorch 示例代码。","2026-09-28 17:49:25","技术",[245,246,247,215],"强化学习","RL","Spinning Up",{"path":249,"title":250,"description":251,"date":252,"categories":243,"tags":253,"draft":185},"\u002Fposts\u002Fspinning-up-rl-intro-2","[译] 深度强化学习入门（二）：强化学习算法的分类","OpenAI Spinning Up 第二部分的全文翻译：无模型与基于模型的 RL、策略优化与 Q-learning 的取舍，以及规划、专家迭代和经验增强等方法。","2026-09-28 10:20:21",[245,246,247,215],{"path":255,"title":256,"description":257,"date":258,"categories":243,"tags":259,"draft":185},"\u002Fposts\u002Fspinning-up-rl-intro","[译] 深度强化学习入门（一）：RL 的关键概念","OpenAI Spinning Up 教程第一部分的全文翻译：智能体与环境、状态与观测、动作空间、策略、轨迹、奖励与回报、价值函数、贝尔曼方程、优势函数，附 MDP 形式化。","2026-09-28 09:29:40",[245,246,247,215],{"path":261,"title":262,"description":263,"date":264,"categories":182,"tags":265,"draft":188},"\u002Fposts\u002Fwhen-machines-come-alive","冰箱之后，生活还会怎样改变？","从厨房里的果蝇想到冰箱、移动互联网与 AI 的多轮调用：AGI 已经到来，只是昂贵的智能会怎样改变生活？","2026-09-25 06:36:22",[266,267,268],"AI","随笔","科技",{"path":270,"title":271,"description":173,"date":272,"categories":173,"tags":204,"draft":188},"\u002Fposts\u002Ftest","test","2026-09-20 06:37:44",{"path":274,"title":275,"description":173,"date":276,"categories":173,"tags":204,"draft":188},"\u002Fposts\u002Fmathematical-foundations-of-llms-1","LLM的数学基础（1）：","2026-09-19 22:59:55",{"path":278,"title":279,"description":173,"date":280,"categories":173,"tags":204,"draft":188},"\u002Fposts\u002Ftowards-maximize-serving-capatity","如何在 offline bench 里最大化 LLM serving system 的吞吐","2026-09-19 05:45:05",{"path":282,"title":283,"description":284,"date":285,"categories":233,"tags":286,"draft":185},"\u002Fposts\u002Fllm-memory-compute-tradeoff","LLM 的 Memory-Compute Trade-off：从 MHA 到 MLA、Sparse 与 Linear Attention","ds4.1f 上线了。它的架构 novel 在哪，便宜和快又是从哪省出来的。","2026-09-11 12:00:00",[235,287,288,289,290],"Attention","MLA","DeepSeek","长上下文",{"path":292,"title":293,"description":294,"date":295,"categories":233,"tags":296,"draft":188},"\u002Fposts\u002Fmarkdown-extensions","Markdown 扩展语法速查","本站 Markdown 支持的扩展语法演示与速查：admonition 提示块、GitHub 仓库卡片、spoiler 遮罩、ABC 乐谱、GP 乐谱、音频播放器、浏览器可运行代码块。","2026-09-01 15:30:00",[297],"博客",{"path":299,"title":300,"description":301,"date":302,"categories":182,"tags":303,"draft":185},"\u002Fposts\u002Fwhy-we-still-need-to-write-a-blog","为什么还需要手写博客？","AI这么强，还要古法写博客吗？","2026-09-01 13:17:53",[297],{"path":305,"title":306,"description":173,"date":307,"categories":173,"tags":204,"draft":188},"\u002Fposts\u002Fllm-serving-1","LLM Serving（一）:以在H200上部署GLM-5.2为例","2026-07-21 06:15:44",{"path":309,"title":310,"description":173,"date":311,"categories":312,"tags":204,"draft":185},"\u002Fposts\u002Finteresting-music-rhythm","有趣的音乐节奏","2026-07-16 12:47:24","音乐",{"path":314,"title":315,"description":316,"date":317,"categories":243,"tags":318,"draft":185},"\u002Fposts\u002Ftowards-monosemanticity","LLM 可解释性（一）：SAE与单语义","读 Anthropic《Towards Monosemanticity》。从难以解释的神经元出发, 讲清superposition为什么让单个神经元混成一团, SAE怎么把多语义的激活分解成单语义特征。","2026-07-12 22:00:00",[235,319,320],"可解释性","SAE",{"path":322,"title":323,"description":324,"date":325,"categories":182,"tags":326,"draft":188},"\u002Fposts\u002Fpiketty-ai-china-k-shaped","用皮凯蒂的镜头看 AI 与中国 K 型社会","一次凌晨的对话整理。从 r>g 出发，看 AI 时代的资本分配、中国的世袭化趋势、体制内的二元劳动力市场，以及一个普通 00 后在一线打拼的真实牌面。","2026-06-24 03:00:00",[267,327,266],"经济",{"path":329,"title":330,"description":173,"date":331,"categories":173,"tags":332,"draft":188},"\u002Fposts\u002Fclaude-code-compact-behavior","cluade code compact 行为","2026-06-23 20:17:42",[333,334,335],"Agent","AI Agent","Claude Code",{"path":337,"title":338,"description":173,"date":339,"categories":243,"tags":340,"draft":185},"\u002Fposts\u002Fclaude-code-settings","Claude Code 配置分享","2026-06-23 14:22:12",[235,334,333,335],{"path":342,"title":343,"description":344,"date":345,"categories":233,"tags":346,"draft":185},"\u002Fposts\u002Fmy-thoughts-on-the-ai-industry","简单聊聊对 AI Agent 行业的理解","写这篇文章的初衷，是我发觉逐渐忘了前两个月的 AI\u002FLLMs 是什么样的能力和情况了。意识中似乎某些在突然加速，但理性和直觉告诉我这不对，作为行内人，我知道这是一个渐进的、累积的过程。我想自己回忆一下，Agent的发展。","2026-06-20 13:14:25",[235,334,333],{"path":348,"title":349,"description":173,"date":350,"categories":243,"tags":351,"draft":185},"\u002Fposts\u002Frotational-positional-encoding","RoPE","2026-06-17 18:31:40",[235,352,349],"Transformer",{"path":354,"title":355,"description":173,"date":356,"categories":243,"tags":357,"draft":185},"\u002Fposts\u002Fintroduction-to-large-language-model","大语言模型入门基础","2026-06-16 22:36:11",[352,235],{"path":359,"title":360,"description":173,"date":361,"categories":233,"tags":362,"draft":185},"\u002Fposts\u002Fthought-about-coding-agent","关于 Coding Agent 使用的遐思","2026-06-03 17:54:20",[333],{"path":364,"title":365,"description":366,"date":367,"categories":243,"tags":368,"draft":185},"\u002Fposts\u002Fabout-ssh-usage","关于SSH的使用——客户端篇","介绍SSH客户端的用法。","2025-06-17 02:44:57",[369],"工具",{"path":371,"title":372,"description":373,"date":374,"categories":233,"tags":375,"draft":185},"\u002Fposts\u002Fopensourcesoftwarestarttutorial","如何开始一个开源项目？","浅谈我对开源项目的认知。","2025-04-17 18:00:00",[376],"开源",{"path":378,"title":379,"description":380,"date":381,"categories":382,"tags":383,"draft":185},"\u002Fposts\u002Fmotivation","渴求动力","有意识的生活并主动地行动会使得生活更美好。","2025-03-05 00:44:57","生活-情感",[267],{"path":385,"title":386,"description":387,"date":388,"categories":382,"tags":389,"draft":185},"\u002Fposts\u002F2024summary","2024年终总结","啊，痛彻的时光；啊，悲怆的人！","2025-01-01 01:00:00",[267],{"path":391,"title":392,"description":393,"date":394,"categories":243,"tags":395,"draft":185},"\u002Fposts\u002Fvps-crash-startup","快速搭建个人VPS","如何快速搭建个人VPS。","2024-12-31 19:00:00",[396],"VPS",{"path":398,"title":399,"description":400,"date":401,"categories":243,"tags":402,"draft":185},"\u002Fposts\u002Fdata-label-software","AI数据（自动）标注软件概述","对市面上目前的AI数据（自动）标注软件进行调研。","2024-12-20 14:00:00",[403,404],"数据标注","深度学习",{"path":406,"title":407,"description":408,"date":409,"categories":382,"tags":410,"draft":185},"\u002Fposts\u002Fself-destruction","生活的崩坏与自我毁灭","是什么在牵引着我？我的行为，我的思想？我的感受？我所处的地方？它们在何处？","2024-12-19 12:00:00",[267],{"path":412,"title":413,"description":414,"date":415,"categories":243,"tags":416,"draft":185},"\u002Fposts\u002Fcs194-196_llmagents","LLM Agents 课程笔记","UCB CS294\u002F194-196 Large Language Model Agents 的课程笔记。","2024-11-30 20:00:00",[333],{"path":418,"title":419,"description":420,"date":421,"categories":382,"tags":422,"draft":185},"\u002Fposts\u002Fwhatsyourworth","价值、位置","你们的生存究竟有何价值？如果毫无价值，你们究竟为何存在？","2024-11-29 02:00:00",[267],{"path":424,"title":425,"description":426,"date":427,"categories":382,"tags":428,"draft":185},"\u002Fposts\u002Fblogrelifever2","重启博客v2","忙里偷闲学习前端，还是决定结合想法手搓一个自己的博客。","2024-11-25 00:00:00",[297],{"path":430,"title":431,"description":432,"date":433,"categories":382,"tags":204,"draft":185},"\u002Fposts\u002Fblogrelife","重启博客","在我的表达欲与现实所需(问题解决记录)驱动下，我决定重写博客。","2024-10-21 20:44:57",{"path":435,"title":436,"description":437,"date":438,"categories":382,"tags":439,"draft":185},"\u002Fposts\u002Frevival","240205 复活","好吧，断更了许久，我又终于回来了，我又终于回来了。","2024-02-05 20:00:00",[267,440,441],"生活","返乡",{"path":443,"title":444,"description":445,"date":446,"categories":382,"tags":447,"draft":185},"\u002Fposts\u002Fclouds","「230717」云","我已经很久忘记了不我已经很少把自己的想法在lofter上发布的。","2023-07-17 20:00:00",[267,448],"散文",{"path":450,"title":451,"description":452,"date":453,"categories":182,"tags":454,"draft":185},"\u002Fposts\u002Fsurvival-course-selection","「230704」生存手册——选课制度","选课制度，选课制度是上海大学十分重要的一项制度。","2023-07-07 20:00:00",[193,455,456],"制度","选课",{"path":458,"title":459,"description":460,"date":453,"categories":182,"tags":461,"draft":185},"\u002Fposts\u002Fsurvival-how-to-do-research","「230707」生存手册——如何做科研","就我个人而言，首先你要确定你是真心想做科研，还是为了那一些。",[193,462,195],"科研",{"path":464,"title":465,"description":466,"date":453,"categories":182,"tags":467,"draft":185},"\u002Fposts\u002Fsurvival-semester-system","「230707」生存手册——三学期制","三学期制短平快，允许我们在很短时间里尝试更多不同的东西。",[193,455,468],"上海大学",{"path":470,"title":471,"description":472,"date":473,"categories":382,"tags":474,"draft":185},"\u002Fposts\u002Fmany-truths-many-existences","「230701」诸多正确 诸多存在","世界上有诸多正确诸多存在因此我们所能接受到的那种人信奉的东西，都是他们心中相信的正确的存在，然而确实在不同的我们可以通过","2023-07-01 20:00:00",[267,475],"哲学",{"path":477,"title":478,"description":479,"date":480,"categories":182,"tags":481,"draft":185},"\u002Fposts\u002Fhow-i-choose-major","「230615」关于我选专业的思路","省流专业选择主要和你的兴趣有关，起码你不能讨厌，而跟你的家境，你是否想要留在上海？","2023-06-18 20:00:00",[193,482,195],"专业选择",{"path":484,"title":485,"description":486,"date":487,"categories":182,"tags":488,"draft":185},"\u002Fposts\u002Fdesire-for-improvement","「230525」保持对提升的渴求","进入大学需要保持对提升的渴求。","2023-05-25 20:00:00",[267,489,195],"成长",{"path":491,"title":492,"description":493,"date":494,"categories":182,"tags":495,"draft":185},"\u002Fposts\u002Fhow-students-rest","「230520」对于学生如何休息的看法","如何正确休息是对更好工作的一个非常有力的支持，而且这是你想做的事情必不可缺的，也就是我们通常俗话说的，当你给到状态好的时","2023-05-20 20:00:00",[193,496,440],"休息",{"path":498,"title":499,"description":500,"date":501,"categories":182,"tags":502,"draft":185},"\u002Fposts\u002Fexam-week-fullness","「230510」考试周的充盈感","不知其他大学如何，但是数的考试之后，期末周的感觉是非常明显的。","2023-05-12 20:00:00",[193,503,440],"考试",{"path":505,"title":506,"description":507,"date":501,"categories":182,"tags":508,"draft":185},"\u002Fposts\u002Fheavy-learning","「230421」沉重的学习","在过去长达12年的教育中，我们已经被养成了这样的习惯，我们不需要去思索知识来源的真实性，直接去接受学校喂给我们的知识就可",[193,509],"学习",{"path":511,"title":512,"description":513,"date":514,"categories":182,"tags":515,"draft":185},"\u002Fposts\u002Fshu-humanistic-care","「230426」shu并不缺乏人文关怀","在我看来上海大学是一座呃有拥有非常多人文关怀的的一所学校，他并不缺乏人关怀。","2023-04-26 20:00:00",[193,468,267],{"path":517,"title":518,"description":519,"date":520,"categories":182,"tags":521,"draft":185},"\u002Fposts\u002Fon-self-study","「230420」关于自学","在就是相关学习上的就是知识方面的自习。","2023-04-20 20:00:00",[193,522,195],"自学",{"path":524,"title":525,"description":526,"date":527,"categories":182,"tags":528,"draft":185},"\u002Fposts\u002Fsurvival-not-learning","「230418」现代生存策略之一-不学习","所谓不学习狭义的，不是去学习学校所传授的内容。","2023-04-18 20:00:00",[193,509,267],{"path":530,"title":531,"description":532,"date":533,"categories":182,"tags":534,"draft":185},"\u002Fposts\u002Fearly-rising","「230410」早起相关","准备写篇跟早起相对的内容。","2023-04-10 20:00:00",[440,535],"作息",{"path":537,"title":538,"description":539,"date":540,"categories":182,"tags":541,"draft":185},"\u002Fposts\u002Fgroup-work-decisions","「230408」小组合作与决策","不知道从什么时候起？我似乎失去了一个团队决策的能力。","2023-04-08 20:00:00",[193,542,543],"协作","决策",{"path":545,"title":546,"description":547,"date":548,"categories":182,"tags":549,"draft":185},"\u002Fposts\u002Fworkflow-rethought","「230402」对工作流的再思索","工作流是什么？工作流是工作学习的一整个啊模块化的，模范化的流程，通过这个流程可以满足我们对学习的绝大多数要求为什么我们需","2023-04-03 20:00:00",[550,551],"效率","工作流",{"path":553,"title":554,"description":555,"date":556,"categories":182,"tags":557,"draft":185},"\u002Fposts\u002Fcopy-homework-carefully","「230401」抄作业应当注意","抄抄作业应当注意以下几点，你不能只是单纯的抄作业，就是抄了就不管了，就是闷着头的只管超，也不去思考他这个到底是什么意思。","2023-04-01 20:00:00",[193,509],{"path":559,"title":560,"description":561,"date":562,"categories":182,"tags":563,"draft":185},"\u002Fposts\u002Fcycle-of-choices","「230330」选择的周期","路上看到了残花，夹杂着落下的雨地被人飘落。","2023-03-30 20:00:00",[267,564],"选择",{"path":566,"title":567,"description":568,"date":569,"categories":182,"tags":570,"draft":185},"\u002Fposts\u002Finfo-channels-experience","「230323」关于信息渠道的个人经历","对我本人的信息渠道得知做一个简单的回顾。","2023-03-23 20:00:00",[193,571,195],"信息渠道",{"path":573,"title":574,"description":575,"date":576,"categories":182,"tags":577,"draft":185},"\u002Fposts\u002Fdiet-under-pressure","「230320」生活节奏压力之饮食","生活节奏这一块是非常的快的。","2023-03-20 20:00:00",[440,578],"饮食",{"path":580,"title":581,"description":582,"date":576,"categories":182,"tags":583,"draft":185},"\u002Fposts\u002Fsemester-gap","「230320」学期gap","因为我们学校的三学期制学期与学期之间的空隙就显得十分狭小，而平凡时期之间的转换。",[193,584],"gap",{"path":586,"title":587,"description":588,"date":589,"categories":182,"tags":590,"draft":185},"\u002Fposts\u002Fart-of-asking-social","「230318」提问和社交的艺术——导生的用处，大学第一课","年入学及入学时就分配导向，是一个非常有助力的资源。","2023-03-18 20:00:00",[193,591,592],"社交","提问",{"path":594,"title":595,"description":596,"date":597,"categories":182,"tags":598,"draft":185},"\u002Fposts\u002Frecommending-and-recommended","「230317」推荐与被推荐","都是较为困难的一个行为，因为这个行为需要对方真心喜欢，并且去实行落实这个行为。","2023-03-17 20:00:00",[193,591],{"path":600,"title":601,"description":602,"date":603,"categories":182,"tags":604,"draft":185},"\u002Fposts\u002Fagainst-weaponized-psychoanalysis","「230316」反对将“精神分析”作为武器","我并不反对分析他人和自己行为原因的这种行为。","2023-03-16 20:00:00",[267,217],{"path":606,"title":607,"description":608,"date":609,"categories":182,"tags":610,"draft":185},"\u002Fposts\u002Flonely-city-future","「230314」未来的孤独城市","在这个寂寞的大多数里面走着，在野外大街道所看到的。","2023-03-14 20:00:00",[267,611],"城市",{"path":613,"title":614,"description":615,"date":616,"categories":382,"tags":617,"draft":185},"\u002Fposts\u002Fsense-of-beauty","「230312」美的感觉","每个时代的人们所具有的美学的感受是并不相同的，这是否可以理解为他们与当前他们所处的时代的社会签订了某种契约，他们一切一切","2023-03-12 20:00:00",[267,618],"美学",{"path":620,"title":621,"description":622,"date":623,"categories":382,"tags":624,"draft":185},"\u002Fposts\u002Fwho-turned-on-light","「230310」谁开的灯","啊，早上5:30起来的时候，分分类垃圾的那位大爷就已经开始在理着了，或许他可能在这之前就变已经开始了工作，所以当我们在早","2023-03-10 20:00:00",[267,448],{"path":626,"title":627,"description":628,"date":629,"categories":243,"tags":630,"draft":185},"\u002Fposts\u002Fblind-people-use-computer","【230309】盲人如何使用电脑？","人并非每个部位都能健康完好，对于有一定缺陷的他们来说，在这个高度信息化的时代，不能使用如电脑、手机等电子设备无疑是会给生","2023-03-09 20:00:00",[631,243,632],"无障碍","思考",{"path":634,"title":635,"description":636,"date":629,"categories":182,"tags":637,"draft":185},"\u002Fposts\u002Freal-way-university-learning","「230309」大学学习的真正方式","大学的学习方式并非和以往的学习方式有什么根本性的不同，我认为在大学之中的学习内容将不会再局限于以前平时呃中学时代那种较为",[193,509,195],{"path":639,"title":640,"description":641,"date":642,"categories":182,"tags":643,"draft":185},"\u002Fposts\u002Fwhat-i-should-do","「计划」思考自己应当做的事","考完了英语思考一些事情。","2023-03-08 20:00:00",[267,644],"规划",{"path":646,"title":647,"description":648,"date":649,"categories":233,"tags":650,"draft":185},"\u002Fposts\u002Fdigital-portrait","【230307】如何描绘数字人物画像","如果让你用数据描绘一个人，你会从哪几个方面描述呢？","2023-03-07 20:00:00",[243,651,632],"数字人",{"path":653,"title":654,"description":655,"date":656,"categories":382,"tags":657,"draft":185},"\u002Fposts\u002Ffreshman-social-psychology","「re：freshman」社交与心理，调节及和解","作为一个曾经作为一个曾经差点抑郁过，或者说差点从心理问题发展为心理疾病的人，哎，我可以这样说，这所学校是对那些尚未脱离自","2023-03-06 20:00:00",[193,658,591],"心理",{"path":660,"title":661,"description":662,"date":663,"categories":182,"tags":664,"draft":185},"\u002Fposts\u002Ffreshman-mbti","「re：freshman」了解自己之mbti","如果我高考之后，从高考的巨大束缚的网络，中途之后在一个相对宽松，肯定一下我们应该思考自己究竟是谁将要干什么？","2023-03-05 20:00:00",[665,658,666],"MBTI","自我认知",1791806753236]