由清华大学研究团队牵头，多所海内外知名高校与研究机构的科研人员共同推出了 OSWorld-Science ：一个面向真实科研场景的计算机使用智能体评测基准及配套基础设施。

文章浏览阅读244次。10.5-10.11 GitHub周榜：13个项目61,769星，环比-12.1%。PS5移植工具AnyPS5（+21,365星）空降第一；Agent席位7席星数反缩水36%；非AI应用接棒，占比57.1%。

2026年9月，硅谷的咖啡还没凉透，一场静默却震耳欲聋的范式革命，早已在百万开发者桌面悄然落地——不是ChatGPT，不是Sora，而是那个曾被当作‘CopilotPlus版’低调存在的Codex，正以肉眼可见的速度，把‘软件工程师’这个称谓，从职业头衔，改写成‘智能体编排师’。 这不是夸张。就在上个月，OpenAI官方宣布：Codex活跃用户突破2370万，其中超68%为专业开发者；而更炸裂的是 ...

文章浏览阅读79次。自动化迁移最容易翻车的地方是：智能体特别擅长宣布自己完成了。所以这个项目先把"完成"变成机器可判定的检查。原文说得很直接：人类唯一必须做的工作，就是把验证搭好，让大规模自主交付变得可能。TUI parity：把 ...

How OpenAI’s Core Experimentation team uses Codex to remove unused code, clean up stale experiments, and improve CI.

Now, anyone can prompt an AI model to construct complex computational pipelines that would otherwise require days or weeks of ...

This video explains how AI agents are transforming the future of software development. It demonstrates that AI will not ...

Infosys, a top Information technology (IT) services firm of India, is looking to hire freshers and grow its Artificial ...

Google has expanded Gemini 4 Argon to API customers and paid Gemini users, following an initial release limited to employees ...

A new open-source package called Varmapack simulates VARMA time series exactly from the first observation, eliminating ...

Python 3.15 is out with explicit lazy imports, UTF-8 as the default encoding, frozendict and sentinel types, the Tachyon ...

When Google DeepMind announced Gemini 4 Argon a few weeks ago, the lab said it would be limiting initial access to the new AI ...