👁 378 次阅读
HackerNews 今日热榜
| rank | title | score |
|---|---|---|
| 1 | Tailscale Traces Database Corruption to 16y/o SQLite WAL-Reset Bug | 875 |
| 2 | DeepSeek V4 Pro 0813 | 807 |
| 3 | Qwen3.8-2.4T | 534 |
| 4 | 2026 Eclipse Webcams | 461 |
| 5 | Delta | 455 |
| 6 | Tim King, AmigaDOS developer, has died | 246 |
| 7 | Principia Mathematica is modern and insightful | 84 |
| 8 | Build Wide, Ship Narrow | 54 |
| 9 | Happy 45th Birthday to the IBM PC and Model F/XT | 45 |
| 10 | Antiqua–Fraktur Dispute | 12 |
GitHub 热门AI项目
| rank | repo | stars | description |
|---|---|---|---|
| 1 | i3T4AN/KADATH | 173 | Evolutionary multi-agent runtime that breeds, evaluates, and |
| 2 | jundizhou/easy-stock | 141 | A 股行情分析与AI智能投研智能体–基于大模型 |
| 3 | vuphongle/oss-pr-reviewer | 39 | AI-powered CLI for reviewing GitHub pull requests, detecting |
| 4 | Leonxlnx/unlazy | 31 | Anti-laziness skill for AI agents. Core: the Depth Tree meth |
| 5 | 0toshigami/orbis-pictus | 26 | A tap-to-explore picture book where an AI draws every page i |
| 6 | yangzc/ainote | 24 | AI agent workflow platform — visual flow orchestration, drag |
| 7 | SodaMem/SodaMem | 22 | Agentic memory infra for AI agents |
| 8 | biuworks/bilibili-digest | 22 | 把 B 站视频变成学习资源的 Chrome 扩展：字幕阅读、双语对照、AI 概览、划词解释和带时间戳的笔记 |
| 9 | z15114664687-dot/hr-onboarding-agent | 21 | Open-source AI-assisted HR onboarding for Feishu/Lark: confi |
| 10 | memoket/memoket-kite | 20 | Memory layer for AI agents with token-efficient, explainable |
Arxiv 最新AI论文
| rank | title | authors |
|---|---|---|
| 1 | AI4AI at Test-Time: Strong-to-Weak Capability Transfer via H | Cheng Qian, Wenting Zhao, Liangwei Yang |
| 2 | Structural Silence: When AI Infrastructure Fails Speakers of | Avijit Roy, Proma Roy |
| 3 | One Frozen Simulator Is Not Enough: Simulator Collapse in Mu | Simon Yu, Nicholas Tomlin, Marwa Abdulhai |
| 4 | VICBench: A Multi-Language Benchmark for Code Vulnerability | Jin Lu, Xuening Han, Yang Zhong |
| 5 | Information Abundance Paradox: Long-Context Training Undermi | Arda Uzunoglu, Benjamin van Du, Daniel Khashabi |
| 6 | Who Thinks Best Depends on How Long You Let Them: Budget-Dep | Rodrigo Guedes , Alison R. Panis |
| 7 | A corpus-specific clinical RAG system matches or outperforms | Praveen Reddy, Charuta Mandke, Suvrankar Datta |
| 8 | Do LLMs Take Care of Their Own? Similarity Signals Can Induc | Akash Kundu, Emanuel Tewolde, Ratip Emin Berk |
| 9 | QV-PIC: Query-Aware Visual Position-Independent Caching for | Yilin Liu, Rui Meng, Wangze Ni |
| 10 | Preference Tree Optimization: Enhancing Goal-Oriented Dialog | Lior Baruch, Moshe Butman, Kfir Bar |
自动采集于 2026-08-13
🔗 相关阅读（你可能还想看）
📄 Gemini Omni 1.2 Flash 发布：视频降价30%、原生15秒，单条成本只贵5分钱
谷歌 11 月 5 日开放 Gemini Omni 1.2 Flash：720p 视频每秒 0.07 美元，比上代降 3
📄 GPT-6 Cyber 发布：121 万美元，948 条通往 root 的路
OpenAI 发布 GPT-6 Cyber：在 Linux 内核上跑出 948 个本地提权利用，整轮 API 成本 12
📄 Anthropic冲刺IPO：170亿营收、8950亿算力承诺、首次在招股书披露递归自我改进风险
📄 Anthropic IPO招股书曝光：单季营收170亿美元背后的8954亿美元算力账单
📄 阿里千问接入 Apple Intelligence：中国版 iPhone 的 AI 时刻来了
📄 苹果AI中国落地靴子落地：千问接入Apple Intelligence，11月4日开测
📄 苹果官宣：Apple Intelligence 国行版落地，阿里千问深度接入 Siri 与系统写作工具
苹果官宣 Apple Intelligence 国行版 11 月 4 日开测、12 月随 iOS 27.3 正式上线。阿
📄 Anthropic 递交 S-1 启动 IPO：Claude 母公司估值剑指 2 万亿美元
📄 Cerebras 公开万亿参数模型部署方案：59 块晶圆跑 3T 参数，推理 2000 tokens/秒