garden
shuffle Random All Bookmarks 3566
Threads
Offline RL suffers from extreme overestimation bias.
Threads
what if copy/paste was smart?
Threads
A computer for your agents https://t.co/WvGSHwBUV6
Code
@zhouluobo: Jev 大家用上没有啊,可太好用了
Books
AI slop applications have made traditional careers pages worse than useless.
Books
An absolutely brilliant interactive textbook Linear Algebra: An Interactive Introduction (3rd editio
Videos
I am working on a crazy video about Deepseek v4.1 Causal Encoder Decoder models.
Threads
there's a therapeutic technique called behavioral activation which can be a little confusing in ther
Newsletters
A full MIT course on visual autonomous navigation.
Code
jev has changed computer-use forever.
Threads
Introducing Parallax: a small, hackable async RL repo with all the bells and whistles of a modern As
Books
We need more examples like this in the open-source RL ecosystem
Code
Build your own harness, folks.
Papers
@ValenTagliabue led this work this over a single fellowship, with @LeonardDung1 and myself advising
Papers
this paper is f*cking brilliant
Courses
I'm a big believer that AI will transform the Earth Sciences, in the way it has Life Sciences. This
Threads
If you suddenly find yourself concerned about AI and would like to transition to working in AI safet
Articles
They had a cool technical post too to train LLMs to predict the future.
Videos
this is insane. i genuinely don't understand why ambitious people aren't shown this lecture before t
Code
Train your own LLM from scratch!
Threads
清华大学和无问芯穹等团队刚刚开源了一篇注定载入史册的重磅论文(已收录于 ICLR 2026,代码已开源):
Code
🔥还有谁?!Jev模型刚出来没几天,NanoJev 已经把整条链路开源了,不是套一层 API,是 0.6B 并行决策模型 + 数据 + 训练管线!
Papers
“In-Context Robot Learning with VLM Agents”
Papers
Introducing the Sakana AI Frontier Intelligence Group 🪷
auto_stories
Select a bookmark to read
Click any item from the list