Blogs
- Time-Series LLMs, Explained with t0-alpha (21:30)
- The Untaught Lessons of RAG Question Parsing: Structure Before You Search (20:00)
- 最新微博：2026 年 7 月 2 日 (18:33)
- 不做结论 (17:40)
- 谁在惩戒那些偏离预期性别角色的人？ (09:00)
- Why Powerful ML Is Deceptively Easy — Part 2 (00:30)
News
- DGX Spark 黑客松线上训练营：4 小时干货，从环境配置到具身智能，手把手教你搭出能跑的 Agent (19:51)
- 让Agent越用越强：AReaL 2.0开源，打造面向自演进智能体的RL基础设施 (19:28)
- 全球首个英伟达含量为0的万亿模型，成了海外开发者的抢手货 (18:56)
- 久坐不动者肌肉线粒体功能出现显著下降 (18:53)
- 科大讯飞举办智能交互生态发布会，三大平台同步升级 (18:45)
- 论坛预告｜夯！AI企业家论坛首发阵容公开，更有千亿级实业巨头突袭 (18:29)
- 天工 3.2 重磅升级：Skywork Tags 上线，给 Agent 一张工牌，邀其加入你的工作群聊 (18:24)
- 来自 Google 的 Android 恶意程序 (18:07)
- 越来越多的儿童使用 AI (16:43)
- 全球昆虫物种可能有 2000 万 (16:29)
- 钉钉A1录音卡入选Gartner AI可穿戴设备报告 (16:03)
- 招募｜量子位RSS’26专题策划 (15:33)
- Cloudflare 推动 AI 公司为内容付费 (15:23)
- 一个广告营销老炮，率先冲刺A股无人驾驶第一股 (15:16)
- 科学家首次利用非生命成分制造出细胞 (14:46)
- 天工AI业务ARR突破8亿美元，向中国首个非BAT10亿美元ARR的AI公司迈进 (14:23)
- 训练世界模型，开始从人类的肌肉和脑子里偷师了 (13:44)
- Anthropic 将移除检测中国用户的秘密代码 (13:41)
Blogs
- 阅读阿伦特如何帮我理解后室和架空文学 (23:35)
- Persistent Latent Memory for Multi-Hop LLM Agents: How a 6G Handover Paper Closes the Agent Cold-Start (23:00)
- What AI benchmarks are not telling you (22:31)
- 申真谞九段受让两子挑战KataGo (22:06)
- For the First Time, a Cell Built From Scratch Grows and Divides (22:00)
- What Can We Do When Memory Becomes the New Bottleneck in Data Engineering? (21:30)
- Announcing the Monetization Gateway: charge for any resource behind Cloudflare via x402 (21:00)
- Content Independence Day, one year on: building the business model for the agentic Internet (21:00)
- Making AI search smarter (21:00)
- Your site, your rules: new AI traffic options for all customers (21:00)
- Build and Run Your Own AI Agent in the Cloud (20:00)
- Unmasking the crawls with Attribution Business Insights (14:00)
- 为什么说 Anthropic 是邪恶的？ (13:59)
- 男性正在遭遇的“系统性”歧视 (09:00)
- Hugging Face and Cerebras bring Gemma 4 to real-time voice AI (08:00)
- 2026上半年研究速递 (08:00)
- Quoting Anthropic (07:58)
- Nano Banana 2 Lite (06:15)
- 台湾：从新竹到台中 (06:02)
- What's new in Claude Sonnet 5 (05:23)
- ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration (02:32)
- The AI Compass (01:39)
- Have your agent record video demos of its work with shot-scraper video (00:54)
- Context Engineering for RAG : The Four Typed Inputs Behind Every RAG Answer (00:30)
- Start building with Nano Banana 2 Lite and Gemini Omni Flash (00:02)
- Vibe Coding新体验 (00:00)
News
- 瑞典法院判决 Google 向比价网站赔偿 15 亿美元 (23:35)
- 索尼 PS 从 2028 年 1 月起不再发售新游戏的光盘版 (22:18)
- Godot 拒绝接受 AI 生成的代码 (21:14)
- LHC 第三次停机维护 (19:08)
- 研究人员将干细胞转化为人类初级卵母细胞 (18:00)
- 中国 2026 年汽车出口量有望达到千万 (15:39)
- arXiv 从康奈尔大学独立 (15:01)
- 美国 Henrico 县请求政府和学校为数据中心节约用电 (14:26)
- 微软发布 WSL 容器的预览版 (13:59)
- 为什么我们需要睡眠 (12:54)
- 美国政府解除对 Claude Fable 5 和 Mythos 5 模型的出口限制 (11:53)
- Claude Code 会悄悄检查用户的系统时区是否是中国 (00:51)
Blogs
- shot-scraper 1.10 (23:10)
- Surviving the Data Science Behavioral Interview (23:00)
- Why Specialization Is Inevitable (22:39)
- How to Maximize Codex Exec Command (21:30)
- Stop Choosing Between Local and Cloud LLMs: A Field Guide to Hybrid Patterns (20:00)
- How ChatGPT adoption has expanded (17:00)
- 下了出租车 (16:19)
- Dario Amodei：AI 开源是伪命题 (11:04)
- 男性有资格认为自己被“歧视”了吗？ (09:00)
- Featuring Every Eval Ever Results on Hugging Face Model Pages (08:00)
- Inside Genebench-Pro (08:00)
- Introducing GeneBench-Pro (08:00)
- Core dump epidemiology: fixing an 18-year-old bug (08:00)
- 刻舟求剑 (08:00)
- HTML table extractor (07:38)
- Count the number of Safari tabs (02:36)
- DiScoFormer: One transformer for density and score, across distributions (02:02)
- How Far Can Classical NLP Go? From Bag-of-Words to Stacking on Spooky Author Identification (01:34)
- 舟山朱家尖亲子自驾4天3晚 (01:15)
- Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding (00:17)
Blogs
- Prompt Engineering Fails Quietly — Prompt Regression Is Why (23:00)
- What Breaks a Cell’s Ribs Can Make It Stronger (22:00)
- I Completed Five Years in Analytics Consulting: 5 Lessons That Changed How I Work (21:30)
- How to Choose Between Small and Frontier Models (20:00)
- Mapping Europe’s AI Workforce Opportunity (15:00)
- 人总会被困在某处。 (10:34)
- 强者心态和弱者心态。 (10:31)
- 大脑充血 Vol.87 (09:44)
- 男性同样也遭受性别歧视？ (09:00)
- HelloGitHub 第 123 期 (08:10)
- Quoting Jon Udell (05:57)
- Hack Your Summer (03:26)
- HP Inc. launches Frontier strategic partnership with OpenAI (01:00)
Blogs
- 在 AI 浪潮下，如何留在牌桌上 (23:51)
- Tail Control: The Counterintuitive Engineering of Reliable Agentic Workflows (23:00)
- 社会学为什么在中国难以发展？｜ 隐说 NO. 27 (21:56)
- I Pitted XGBoost Against Logistic Regression on 358 Matches. The Boring Model Won. (21:00)
- AI 时代的数学研究 (18:45)
- 暑假 (11:01)
- 这一周莫比乌斯环转到了哪里？ (09:00)
- AI见闻之政策 (01:31)
- AI见闻之数据 (01:26)
Blogs
- 为什么我只跟练帕梅拉？ (23:24)
- We Built a Routing Layer to Cut Our AI Costs. It Broke the Product. (23:00)
- How to Build a Powerful LLM Knowledge Base (21:00)
- Je déteste la vi(ll)e (16:40)
- 您受累，搞个大新闻儿！ (09:00)
- AI的浪潮 (08:00)
- Quoting Dean W. Ball (06:25)
- Quoting Timothy B. Lee (05:15)
- 理解 Skill —— 读《图解 Skill》 (03:04)
- What happened after 2,000 people tried to hack my AI assistant (02:33)
- Incident Report: CVE-2026-LGTM (01:58)
- Quoting OpenAI (01:10)
- From Local LLM to Tool-Using Agent (00:30)