👁 154 次阅读
HackerNews 今日热榜
GitHub 热门AI项目
| rank | repo | stars | description |
|---|---|---|---|
| 1 | 0xsline/awesome-deepseek-harness | 661 | DeepSeek Harness (DSH) ecosystem: curated plugins, tools, an |
| 2 | seyedehsanhadi/sloptrim | 150 | A local detector for AI-writing patterns. Scores every prose |
| 3 | vuphongle/oss-pr-reviewer | 109 | AI-powered CLI for reviewing GitHub pull requests, detecting |
| 4 | thiientv/godmode | 90 | Production-grade Agent Skills for AI coding agents—composabl |
| 5 | memoket/memoket-kite | 72 | Memory engine for AI agents with token-efficient, explainabl |
| 6 | biuworks/bilibili-digest | 70 | 已上架 Chrome 商店 🚀 把 B 站视频变成学习资源的浏览器扩展（Chrome / Edge）：字幕阅读、双语对照 |
| 7 | z15114664687-dot/hr-onboarding-agent | 50 | Open-source AI-assisted HR onboarding for Feishu/Lark: confi |
| 8 | Greninja9257/LabLLM | 40 | A native macOS lab for teaching tiny language models to thin |
| 9 | zimodzh/dsh-plugin-dev-skills | 34 | An Agent Skills skill for developing DeepSeek Harness (DSH) |
| 10 | diyanbogdanov/emend | 34 | Verified AI dependency migrations. Finds which of your lines |
Arxiv 最新AI论文
| rank | title | authors |
|---|---|---|
| 1 | Split the Labor: Separating Evidence Interpretation from Dec | Zhelun Wu |
| 2 | Whose doctor does the AI recommend? An algorithm audit of re | Syeda Anshrah G, Mirza Samad Ahm |
| 3 | LLMs Don’t Pay for the Jump | Paras Balani, Subhrakanta Pan |
| 4 | Wrong but Useful: Trajectory Value Beyond Answer Correctness | Chih-Hsuan Yang, Anjir Ahmed Cho, Cheng-Hau Yang |
| 5 | A Four-Axis Trustworthiness Benchmark for LLM-as-Judge in Pr | Dipankar Sarkar |
| 6 | AnchorBench: A Multi-Pathway Benchmark for the Anchoring Eff | Yiderigun Borji, Alexander Herma, Christian Cyron |
| 7 | Seeing Red, Thinking Bad: Color Bias in Vision Language Mode | Kohsuke Ide, Ryousuke Yamada, Yoshihiro Fukuh |
| 8 | SimpleOPD: Simple Tokenizer-Agnostic On-Policy Distillation | Haonan He, Haodi Lei, Yun Luo |
| 9 | Grounding Without Corrective Control: Truth-Tracking Profile | Brett Reynolds |
| 10 | MathForm: Scaling Mathematical Autoformalization with Knowle | Lushi Pu, Weiming Zhang, Xinheng Xie |
自动采集于 2026-08-17
🔗 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