👁 1,430 次阅读
HackerNews 今日热榜
GitHub 热门AI项目
| rank | repo | stars | description |
|---|---|---|---|
| 1 | squall01337/mixamo-llm-mocap | 129 | Turn any video into a Mixamo-rig animation – GVHMR estimator |
| 2 | turkiyeyapayzekaakademisi/llm-rag-memory-ai-agents | 109 | |
| 3 | zorost/AI-Engineering-Lab | 98 | A free, self-paced 24-week AI engineering course: Python, ma |
| 4 | elie222/botdirectory.ai | 97 | Open-source directory of agent-bot prompts for Grok Bot, Rak |
| 5 | zachahn/vomit | 94 | Clean up Claude 5’s token vomit with a separate LLM. Save yo |
| 6 | only-cli/oc | 70 | Turn any website into a compact CLI tailored for AI agents. |
| 7 | Vendredi218/awesome-ai-harness | 68 | The model is the engine; the harness is the car. Curated kno |
| 8 | wtfzambo/speak-like-you-eat | 60 | Translate AI garbage to human language. A Pi package for pla |
| 9 | Greninja9257/LabLLM | 59 | A native macOS lab for teaching tiny language models to thin |
| 10 | abligail/narralume | 50 | Open-source AI-assisted writing studio for long-form fiction |
Arxiv 最新AI论文
| rank | title | authors |
|---|---|---|
| 1 | G-CARL: Grounded Checklist-Aligned Reward Learning for Patie | Shiao Xie, Siyu Chen, Jianwei Lv |
| 2 | An Agentic Approach for Active Data Collection, Travel Behav | Narges Ahmadi, Yubo Jiao, Jônatas Augusto |
| 3 | Inducing Task Models from Computer-Use Traces | Yucheng Jiang, Zora Zhiruo Wan, Ruishi Chen |
| 4 | AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design f | Yizhe Chi, Wenyi Li, Deyao Hong |
| 5 | Phantom Gains: Auditing Self-Improvement Against a Measured | Cheng Xu, Nan Yan, Liming Chen |
| 6 | Inject, Align, Recover: Staged Post-Training for Retrieval-F | Qian Kou, Xiaofeng Shi, Xiaosong Qiu |
| 7 | Break It Down, Pass It On: Cross-Task Skill Transfer in LLM | Yiyang Feng, Biddut Sarker B, Niranjan Balasu |
| 8 | Daedalus-150M: A Convolution-Attention Hybrid Designed for C | Christos Koutsi |
| 9 | ContractScrub: A benchmark for final review of legal contrac | Yejin Bang, Kirsty Fielding, Brandan Oliver |
| 10 | MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use | Mengru Wang, Haozhe Luo, Zhenqian Xu |
自动采集于 2026-08-21
🔗 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