LogoThread Easy
  • 探索
  • 撰写 Thread
LogoThread Easy

您的一体化 Twitter 线程助手

© 2025 Thread Easy All Rights Reserved.

探索

最新在前,按卡片方式浏览线程

开启时会模糊预览图,关闭后正常显示

Cursor的 Composer1 模型限免了
 大家可以用用看 是我最近主力模型 优势就是快  几乎不用等就完成了 准确率还行 当然有些场景有点智障 换替补GPT5就搞定了

Cursor的 Composer1 模型限免了 大家可以用用看 是我最近主力模型 优势就是快 几乎不用等就完成了 准确率还行 当然有些场景有点智障 换替补GPT5就搞定了

独立开发者 自由职业 作品 - 简单简历 https://t.co/xMu5JFIGnr 五分钟打造程序员的金牌简历 课程 - 慕课网精英讲师 https://t.co/NTyFFrvHwL 经历 - 不上班的1000天 https://t.co/bonuLQCCsY 视频 - https://t.co/aQYLgujIyC

avatar for Viking
Viking
Fri Nov 07 14:29:44
How much doe?

How much doe?

Founder | Author | Speaker Building @beltstripe. Healtech/EdTech/Agric I'm Not The Man Of Your Dreams. Your Imagination Wasn't This Great.

avatar for Sani Yusuf
Sani Yusuf
Fri Nov 07 14:28:57
RT @kimmonismus: The worlds smartest AI agent system is
1) opensouce and- weight
2) from china

Let that sink in.

RT @kimmonismus: The worlds smartest AI agent system is 1) opensouce and- weight 2) from china Let that sink in.

Co-founder & CEO @HuggingFace 🤗, the open and collaborative platform for AI builders

avatar for clem 🤗
clem 🤗
Fri Nov 07 14:19:19
The paper:

https://t.co/zYNA51w6fv

2/2

The paper: https://t.co/zYNA51w6fv 2/2

Research Scientist @meta (FAIR), Prof. @Unige_en, co-founder @nc_shape. I like reality.

avatar for François Fleuret
François Fleuret
Fri Nov 07 14:18:03
There is a paper from 2017 that introduced a trick that I love but never seen used.

Consider two linear layers f and g that you initialize with the same parameters, and then you use

h(x)=f(relu(x))+g(-relu(-x))

Then at initialization, h is linear!

1/2

There is a paper from 2017 that introduced a trick that I love but never seen used. Consider two linear layers f and g that you initialize with the same parameters, and then you use h(x)=f(relu(x))+g(-relu(-x)) Then at initialization, h is linear! 1/2

Research Scientist @meta (FAIR), Prof. @Unige_en, co-founder @nc_shape. I like reality.

avatar for François Fleuret
François Fleuret
Fri Nov 07 14:18:02
There is a paper from 2017 that introduced a trick that I love but never seen used.

Consider two linear layers f and g that you initialize with the same parameters, and then you use

h(x)=f(relu(x))+g(-relu(-x))

Then at initialization, h is linear!

1/2

There is a paper from 2017 that introduced a trick that I love but never seen used. Consider two linear layers f and g that you initialize with the same parameters, and then you use h(x)=f(relu(x))+g(-relu(-x)) Then at initialization, h is linear! 1/2

The paper: https://t.co/zYNA51w6fv 2/2

avatar for François Fleuret
François Fleuret
Fri Nov 07 14:18:02
  • Previous
  • 1
  • More pages
  • 561
  • 562
  • 563
  • More pages
  • 2117
  • Next