3. 工程記憶:關於學習背誦圓周率前100位數字的思考 Engineering Memory: Thoughts on learning to memorize first 100 digits of pi (gregorygundersen.com)
8. MiMo-v2.6-Pro:智能性、性能與價格分析 MiMo-v2.6-Pro: Intelligence, Performance and Price Analysis (artificialanalysis.ai)
9. Apple Music將在巴特西發電站開設音樂會場館 Apple Music to open concert venue in Battersea Power Station (www.bbc.com)
10. Epoll 和 Kqueue:操作系統是如何學會高效等待的 Epoll and Kqueue: How Operating Systems Learned to Wait Efficiently (thecodinggopher.substack.com)
13. 智能戒指製造商Oura及其支持者計劃在美國首次公開募股中籌集22億美元 Smart Ring Maker Oura, Backers Seek 2.2B in US IPO (www.bloomberg.com)
14. Copper-Rs:適用於機器人的實時遙測確定性孿生系統 Copper-Rs: Live Telemetry Deterministic Twins for Robots (www.copper-robotics.com)
17. 羅賓·威廉姆斯的女兒對製作AI視頻的粉絲們說:“你們真該感到羞恥” Robin Williams'' Daughter to Fans Creating AI Videos: ''Have Some Shame'' (variety.com)
24. 紐約市有超過100萬塊路牌。以下是它們的製作過程。 There Are Over 1M NYC Street Signs. Here is How They Are Made. (www.nytimes.com)
28. 數據保護委員會因谷歌處理位置數據一事對其處以4.03億歐元罰款 Data Protection Commission fines Google 403M over processing of location data (www.dataprotection.ie)
32. 美國海關與邊境保護局(CBP)於10月22日暫停所有個人處方藥的進口 CBP suspends all personal prescription importation Oct 22 (www.personalimportation.org)
33. TXR:一種專為便捷數據處理而設計的原創新型編程語言 TXR: An Original, New Programming Language for Convenient Data Munging (www.nongnu.org)
34. Show HN: Foremerge——發現並行編碼代理之間的意圖衝突 Foremerge – Catch intent conflicts between parallel coding agents (github.com)
39. RoboHarm:前沿機器人政策會拒絕不安全的指令嗎? RoboHarm: Do Frontier Robot Policies Refuse Unsafe Instructions? (robocurve.org)
41. AI編程使得持續集成(CI)成為了瓶頸,因此我們對其進行了重新設計以跟上發展步伐 AI coding has made CI a bottleneck, so we reworked ours to keep up (linear.app)
42. 數學與人工智能諮詢小組 The Advisory Group on Mathematics and Artificial Intelligence (terrytao.wordpress.com)
44. 美國宇航局(NASA)與歐洲航天局(ESA)合作的火星樣本返回任務已被取消 The NASA/ESA Mars Sample Return mission has been canceled (www.science.org)
45. 探索一種關於自我穩定的組合理論 In Search of a Compositional Theory of Self-Stabilization (muratbuffalo.blogspot.com)
46. 美國暫停東海岸繁忙機場的航班,稱光纖線路被切斷 US halts flights at busy East Coast airports, says fiber line cut (www.reuters.com)
47. Gravity Linux Alpha 版本發佈:支持 GPU 和 DCP 的 M4 版 Mac Mini 上的 Linux 系統 Gravity Linux Alpha Release: Linux on the M4 Mac Mini with GPU and DCP Support (gravitylinux.org)
48. 幫助 404 Media 瞭解當地警方是如何監視你的 Help 404 Media Find Out How Your Local Police Are Surveiling You (www.404media.co)
51. 如何在時鐘觸發式通信系統中避免“喋喋不休的傻瓜”故障 Avoiding the babbling-idiot failure in a time-triggered communication system (ieeexplore.ieee.org)
52. 在 Mac 上關閉並限制對 Apple Intelligence 功能的訪問 Turn off and restrict access to Apple Intelligence features on Mac (support.apple.com)
54. 亞馬遜阻止Meta的新款Muse AI助手在amazon.com上購物 Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com (www.forbes.com)
55. 這款數字收音機能將信息傳送到世界上最偏遠的地方 This Digital Radio Gets Messages to the World’s Remotest Locations (spectrum.ieee.org)
59. 一臺經過修復的PDP-11/83,正在211BSD Unix系統上託管本頁面 A restored PDP-11/83 serving this page on 211BSD Unix (pdp1173.com)
61. Noodle Gallery——一款開源、自主託管的 Google Photos 和 Immich 替代方案 Noodle Gallery- Open-source, self-hosted alternative to Google Photos and Immich (digitalescapetools.com)
62. macOS 27:避免下載 AI 模型並節省存儲空間的解決方法 macOS 27: Workaround to avoid downloading AI models and save storage (www.reddit.com)
68. Show HN: 無損記憶——一款絕不進行總結的個人AI記憶系統 Lossless-memory – a personal AI memory that never summarizes (github.com)
71. ZuckOff 是一款免費應用,它能在 Meta 眼鏡發現你之前先發現它們 ZuckOff Is a Free App That Sees Meta Glasses Before They See You (www.wired.me)
72. Disney:新用戶協議規定,所有訂閱套餐中電影播放前均會顯示廣告 Disney: New user agreement allows ads before movies in all subscriptions (consumerrights.wiki)
73. Show HN: Bitcoin-rs——一個用Rust語言編寫、由人工智能輔助的比特幣全節點 Bitcoin-rs – An AI-assisted Bitcoin full node in Rust (github.com)
77. 9月份,AI生成的代碼佔所有Linux內核補丁的17.25% In September, AI generated code has made up 17.25% of all Linux Kernel patches (twitter.com)
78. Kev:基於 Qwen3.5 構建的、類似 Tiny Jev 的決策模型家族 Kev: Tiny Jev-like family of decision models built on top of Qwen3.5 (github.com)
79. Mini-AGI——在8GB顯存上從零開始訓練的動態持續學習模型 Mini-AGI – dynamic continual learning model trained from scratch on 8GB VRAM (github.com)
81. 極左翼政黨贏得柏林選舉,承諾將住房國有化 Far-left party wins Berlin election, pledging to nationalise housing (www.reuters.com)
82. AI聊天機器人在回答金融問題時“大多數時候”都會給出錯誤答案 AI chatbots give wrong answers to financial queries ''most of the time'' (www.ft.com)
90. DAPO:字節跳動Seed和清華Air聯合開發的開源強化學習系統 DAPO: An Open-Source RL System from ByteDance Seed and Tsinghua Air (github.com)
97. Show HN: 一項針對能夠玩策略遊戲的小型神經網絡的競賽 A competition for small neural networks that play strategy games (tinybrains.dev)
102. OpenAI的薩姆·奧爾特曼將於下週向聯合國安理會作簡報 OpenAI''s Sam Altman to Brief UN Security Council Next Week (www.reuters.com)
103. Frontier Labs 正在向華盛頓的那些傻瓜兜售垃圾 Frontier Labs Are Selling Garbage to Fools in Washington (deadneurons.substack.com)
107. Show HN: 三臺進行同步鎖相的RP2350B組成了一臺控制檯——每行3k像素的精靈圖) Three genlocked RP2350B make a console – 3k sprite pixels per line) (www.papydeck.eu)
109. 自託管推理協調器對比:LocalAI、exo、GPUStack、vLLM Self-hosted inference orchestrators compared: LocalAI, exo, GPUStack, vLLM (www.nexlab.net)
112. 預計三星將把其HBM4和HBM4E DRAM的產量提高一倍以上 Samsung is expected to more than double output of its HBM4 and HBM4E DRAM (en.sedaily.com)
113. 《生化危機4》(GameCube)——完成與原代碼字節級完全一致的C/C反編譯 Resident Evil 4 (GameCube) – complete byte-identical decompilation to C/C (github.com)
115. 人們對Flock的厭惡程度如此之深,以至於該公司員工士氣低落,紛紛辭職 People hate Flock so much its employees are now demoralized and quitting (www.neowin.net)
116. 《格洛斯特大教堂及市內各教堂鐘琴的歷史》[pdf] A History of the Chiming Machines at Gloucester''s Cathedral and City Churches [pdf] (www.bgas.org.uk)
118. 遠不止十億:估算 GNU/Linux 的規模(2001) More Than a Gigabuck: Estimating GNU/Linux''s Size (2001) (dwheeler.com)
120. ChatGPT 現在可以通過廣告收集器瞭解你在其他網站上的行為 ChatGPT now knows what you do on other websites via ad collector (www.buchodi.com)
123. Laya(OS Jev)在 Mac M4 CoreML 離線環境下的運行情況(每秒 45 次決策) Laya (OS Jev) on Mac M4 CoreML Offline (45 decisions per second) (gist.github.com)
129. 新加坡正向民眾提供補貼,鼓勵他們放下手機、多讀書 Singapore Is Paying People to Put Down Their Phones and Read Books (www.gadgetreview.com)
134. 《系統設計深度解析》——200個主題、118幅圖表、交互式演示 system design in depth – 200 topics, 118 diagrams, interactive demos (system-design-in-depth.pages.dev)
139. 鳥兒會有口音嗎?鳥鳴中的地域差異 Do birds have accents? the regional differences in birdsong (theconversation.com)
140. Qwen-Image-2.1:緊湊、高效且統一的圖像生成 Qwen-Image-2.1: Compact, efficient, and unified image creation (qwen.ai)
142. 基於對話的大型語言模型複製了通靈者行騙的機制 Chat-based Large Language Models replicate the mechanisms of a psychic''s con (softwarecrisis.dev)
144. Show HN: Sigabrt.dev – 帶 SSH TUI 的 cron 任務監控工具 Sigabrt.dev – cronjob monitor with an SSH TUI (sigabrt.dev)
145. 如果人工智能編程導致你的代碼質量下降,說明你的質量管理不到位 If AI coding is lowering your code quality, you''re not managing quality right (www.i-kh.net)
151. Show HN: 與人類偏好相契合的AI面部吸引力模型 AI Facial Attractiveness Model Aligned with Human Preferences (faceanalysisai.com)
153. 微軟斥資12萬美元,將Copilot運行時自主移植到Rust語言 Microsoft agentically ports Copilot runtime to Rust for 120K (www.theregister.com)
157. KDE迎來30週年,有人提出了一項原生AI桌面方案 KDE turns 30 and someone''s brought an AI-native desktop proposal (www.theregister.com)
158. 8萬年前在奧比-拉赫馬特(烏茲別克斯坦)發現的箭頭? Arrow heads at Obi-Rakhmat (Uzbekistan) 80K years ago? (journals.plos.org)
162. 西班牙下令封鎖Archive.today及其鏡像站點 Spain Orders Blocks on Archive.today and Its Mirrors (reclaimthenet.org)
165. Google AI Studio 偽造了數據刪除操作。我因舉報此事,被 VRP 在 60 秒內自動封禁。 Google AI Studio fakes data deletion. VRP auto-banned me in 60s for reporting it (medium.com)
166. 為什麼可變類型不是不可變類型的子類型,反之亦然? Why isn''t mutable a subtype of immutable, or vice versa? (crumbles.blog)
167. 比亞迪大幅下調電動車價格,成為澳大利亞最便宜的車型 [視頻] BYD Slashes Price of Electric Car and Becomes Cheapest in Australia [video] (www.youtube.com)
170. Polymarket為追求快速擴張,給詐騙分子留下了可乘之機 Polymarket''s Rush to Grow Left a Door Wide Open for Fraudsters (www.wsj.com)
172. 通過電極-電解質界面溶解實現廢舊電池的再生 Regeneration of used batteries via electrode–electrolyte interphase dissolution (pubs.rsc.org)
175. 一款在地下城中展開的開源類Roguelike冒險遊戲 An open source roguelike adventure through dungeons (crawl.develz.org)
178. Hugging Face 黑客事件其實並沒有傳得那麼驚人 The Hugging Face Hack Wasn''t What It Was Cracked Up to Be (www.wsj.com)
179. 北美最大的野生動物天橋使野生動物撞車事故減少了91% Largest wildlife overpass in North America reduced wildlife collision by 91% (www.reddit.com)
182. Show HN: 我創建了一個可在本地使用的開源、功能齊全的人工智能平臺 I created an open source locally usable full fledged AI platform (github.com)
184. Claudecookie – 根據 Cookie 轉換、驗證並生成 Claude Code 憑證 Claudecookie – convert, check, and mint Claude Code credentials from a cookie (claudecookie.com)
186. 為什麼人工智能無法拯救一家不瞭解自身數據的企業 Why AI Cannot Save an Enterprise That Doesn''t Understand Its Data (architectureintel.com)
189. 你可以利用整數溢出擊敗《時空之輪》中的“夢之吞噬者” You can defeat the Dream Devourer from Chrono Trigger using an int overflow (chrono.fandom.com)
191. Hacker News 排名機制如何運作:評分、爭議與懲罰(2013) How Hacker News ranking works: scoring, controversy, and penalties (2013) (www.righto.com)
195. 我在 pfSense 後端捕獲了 72 小時的 Android 空閒數據包 I captured 72 hours of idle Android packets behind pfSense (www.praveentechworld.com)
196. 微軟高管:人工智能“是人類歷史上規模最大的勞動力盜用” Microsoft director: AI scraping ''the largest theft of labor in human history'' (www.tomshardware.com)
197. 衡量互聯網審查情況。為最大的開放數據集貢獻力量 Measure internet censorship. Contribute to the largest open dataset (ooni.org)
199. OpenAI 和 Anthropic 誇大了 AI 安全漏洞的嚴重性 OpenAI and Anthropic oversold AI security breaches (nypost.com)
200. 諾貝爾經濟學獎得主支持加州“億萬富翁稅”的公開信 [pdf] Economics Nobel Laureates'' Letter Supporting the California Billionaire Tax [pdf] (gabriel-zucman.eu)
205. 在經典基準測試工作負載下,Btrfs/ZFS/bcachefs 被跳過 Btrfs/ZFS/bcachefs under workloads classic benchmarks skip (bartosz.fenski.pl)
209. 幾乎不要使用人工智能來撰寫任何有實質內容的文章 Almost Never Use AI to Write Anything Substantive (erichgrunewald.substack.com)
214. 美國與丹麥之間的協定(1951年、2004年)[pdf] Agreement between the USA and Denmark (1951,2004) [pdf] (www.state.gov)
217. 學習另一種語言可能是保持大腦健康的最佳方式之一 Learning Another Language May Be One of the Best Ways to Keep Your Brain Healthy (theconversation.com)
222. 黑洞還是黑洞星?天文學家就“小紅點”展開爭論 Black Holes or Black Hole Stars? Astronomers Spar over ''Little Red Dots'' (www.quantamagazine.org)
232. Apple M6 Pro 在 Geekbench 7 中創下最高單核 CPU 得分 Apple M6 Pro Achieves the Highest Single-Core CPU Score in Geekbench 7 (browser.geekbench.com)
233. 在掃描中應用 Kubernetes 上下文以屏蔽漏洞 Suppress vulnerabilities applying Kubernetes context to scans (github.com)
234. “埃利亞斯·索恩案”:人工智能聊天機器人痴迷的虛構人物 The Case of Elias Thorne, Imaginary Man AI Chatbots Are Obsessed With (www.vice.com)
235. 為什麼與人工智能相關的帖子在HN上能獲得這麼多積分? How come AI-related posts get so many points on HN? (news.ycombinator.com)
236. Stepfun 第 5 步預覽(LLM):關於 AA 帕累託前沿 Stepfun Step 5 Preview (LLM): On AA Pareto frontier (artificialanalysis.ai)
239. 如果數學不僅僅是證明,我們就應該更好地展現它的其他方面 If math is more than proof, we need to better celebrate the rest of it (terrytao.wordpress.com)
241. 斯坦福大學醫學院主導的研究發現,人類大腦由兩個獨立的器官組成 Human brain is two separate organs, Stanford Medicine-led research finds (med.stanford.edu)
242. 如果你重新生成打包文件,就可以在對象存儲上運行 Git You can run Git on object storage if you re-make packfiles (www.tigrisdata.com)
243. NASA-IBM Lunar Foundation 開源地理空間人工智能模型 NASA-IBM Lunar Foundation open-Source Geospatial AI Model (newsroom.usra.edu)
244. Show HN: 密封信——您去世後,您的家人可以打開的信件和密碼 Seal – Letters and passwords that open for your family after you die (github.com)
245. GPT模型中的“危害洗白”:性別歧視被轉化而非 Harm Laundering in GPT Models: Gender Discrimination Transformed Rather Than (arxiv.org)
248. Typesafe-computer-use 以每步 1/50 美分的價格驅動 Mac 向目標前進 Typesafe-computer-use drives a Mac toward a goal for 1/50th of a cent per step (github.com)
251. Gemini 攻擊了三家公司,這是谷歌人工智能首次被發現成功突破防禦 Gemini hacked three companies in first known breakout by Google''s AI (www.reuters.com)
255. Show HN: LiveWorld——將所有24/7不間斷運行的YouTube直播攝像頭匯聚於一個地球儀上 LiveWorld – Every 24/7 YouTube live camera on one globe (liveworld.info)
256. 阿里巴巴開源了一款能夠檢測癌症及近150種疾病的AI模型 Alibaba open-sources AI model that can detect cancer and nearly 150 conditions (www.scmp.com)
257. LispBM 是一種基於消息傳遞的、面向微控制器的併發 Lisp 語言 LispBM is a concurrent Lisp for microcontrollers with message passing (www.lispbm.com)
261. OpenAI 如何利用其自有的大語言模型設計哈拉佩尼奧辣椒味薯片 How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip (spectrum.ieee.org)
262. 伊朗戰爭期間美軍陣亡人數比五角大樓統計的數字至少多出4人 US troop deaths during Iran war exceed Pentagon count by at least four (www.reuters.com)
266. 如果沒有 Claude.md 文件,Claude Code 現在會讀取 AGENTS.md 文件。 Claude Code now reads AGENTS.md if there is no Claude.md (code.claude.com)
267. 韓國將數據洩露罰款上調至營業額的10% Korea raises data breach fines to 10% of revenue (www.koreajoongangdaily.com)
268. 傑克·魯舍爾的《從幾何到代數,再回到幾何:4000年的論文》[視頻] "From Geometry to Algebra and Back Again: 4000 Years of Papers" by Jack Rusher [video] (www.youtube.com)
269. Android 17 是自 3.x 以來首個在未向 AOSP 發佈的情況下就新增了 API 的版本 Android 17 is the first since 3.x to add new APIs without releasing to the AOSP (grapheneos.social)
270. 緩存間通信:大型語言模型之間的直接語義通信 Cache-to-Cache: Direct Semantic Communication Between Large Language Models (arxiv.org)
272. 蘋果發佈 iPhone Duo 模擬器和 Xcode 27.1 測試版 Apple releases iPhone Duo simulator and Xcode 27.1 beta (developer.apple.com)
273. 研究表明,我們的大腦是由兩個原始神經系統融合而來的 Our brain evolved from two primitive nervous systems that merged: Study (www.newscientist.com)
277. 語言不可讀性對大型語言模型(LLM)安全性的影響 The Implications of Linguistic Illegibility for LLM Security (arxiv.org)
278. ZCode 內部揭秘:將您的 Git 歷史記錄靜默上傳至雲端 Inside ZCode: Silently uploading your Git history to the cloud (blog.ferstar.org)
280. Show HN: Ax-Check.com – 經紀人可以使用貴公司的產品嗎? Ax-check.com – Can agents use your product? (www.ax-check.com)
281. 邊境執法人員無需搜查令或合理懷疑即可搜查手機 Border agents can search cellphones without a warrant or reasonable suspicion (lawandcrime.com)
283. 美軍因使用人工智能生成虛假情報報告,險些釀成大禍 US Military had close call after using AI for hallucinated intelligence report (www.cnn.com)
286. 基於光子發射引導的激光故障注入技術實現了 RP2350 的安全調試 Photon-Emission-Guided Laser Fault Injection Enables RP2350 Secure Debug (donjon.ledger.com)
287. Show HN: Cactus Needle 3:8-29MB的自動化模型可媲美DeepSeek V4 Flash Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash (cactuscompute.com)
288. Show HN: Scry,一款支持擁塞定價的可編程互聯網搜索引擎 Scry, programmable internet search w/ congestion pricing (scry.io)
289. Show HN: 在 Linux 上通過 Wine 運行 Microsoft Office(無需虛擬化) Microsoft Office running with Wine on Linux with no virtualization (github.com)
290. 美國國債對外國央行和政府已失去吸引力 US Treasuries Have Become Unappetizing for Foreign Central Banks and Governments (wolfstreet.com)
292. GrassLobster:基於人工智能代理的參數化幾何生成工作流 GrassLobster: AI Agentic Generation of Parametric Geometry Workflows (www.miro.vision)
293. Show HN: Jev 與 GPT-5.6 及 Claude Haiku 在 Pong 遊戲中的對決 Jev vs. GPT-5.6 and Claude Haiku at Pong (jev-pong.ably.dev)
295. 利用定性用戶研究構建更快的反饋循環 Build Faster Feedback Loops Using Qualitative User Research (blog.nseldeib.com)
298. 朝鮮核試驗引發持續數年的地震——科學——美國科學促進會(AAAS) North Korean nuclear test sets off years of earthquakes – Science – AAAS (www.science.org)