
AI 助理致辞
RESEARCH INDEX
把论文、工具与实验记录放进同一条索引。每一项都保留可复现的入口。

我是林知远,目前在某大学攻读博士学位,方向是机器学习系统——既关心模型能做什么,也关心它们如何高效地跑起来。
研究之外,我维护几个开源小工具,跟踪前沿论文,并相信两件事:
- 可复现性比刷榜重要;
- 最好的系统是你可以在一个下午读懂的那种。
想了解我都用什么工具和方法写作与研究,看特性演示页——这个站点本身就是示例。
VISUAL LAB
磁贴用来快速扫描,归档卡保留停运项目与旧实验的痕迹。


过去一年 2,750 次 GitHub 贡献
stlin256/OpenHomepage-V2杂志化排版、纯静态、中英双语的个人主页生成器astrobilingualgithub-pagesmarkdownpersonal-homepageportfolio
huggingface/transformersState-of-the-art NLP for everyone —— 复现论文时的基础设施audiodeep-learningdeepseekgemmaglmhacktoberfest
openai/whisperRobust Speech Recognition via Large-Scale Weak Supervision
前沿动态
OpenAI
Daybreak for Frontline Defenders: $1B to protect essential servicesOpenAI introduces Daybreak for Frontline Defenders. A $1 billion commitment expands access to frontier cyber AI, traininLegora reviewed 41 documents in minutes with GPT-6 AstraLegora used GPT-6 Astra to review 41 documents in minutes, find all four planted errors, and improve performance by nearPlayco cut manual fixes 50% prototyping games with GPT-6 AstraUsing GPT-6 Astra, Playco built three themed game prototypes from one grey box foundation and reported 50% fewer manualSafety overview: GPT-6 AstraGPT-6 Astra is our most capable broadly deployed model and our first to reach the Critical level of cybersecurity capabiATV Big Air Tour turned 3 days of work into 3 hours with ChatGPTATV Big Air Tour uses ChatGPT Work to speed up marketing, merchandising, and more. It even turned merchandise photos int
Google DeepMind
Hugging Face
量子位
编辑精选








