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 v2.0 released

Features
- Run up to eight agents in parallel
- Add Composer agentic coding model (4x faster)
- Promote Browser to GA and embed in-editor
- Make sandboxed terminals GA and default on macOS
- Offer Cloud Agents with 99.9% reliability and instant startup
- Add Team Commands for centrally managed team rules
- Allow building plans in background with another model
- Improve prompt UI: show files as inline pills and let Agent self-gather context

Fixes
- Fix memory leaks and reduce overall memory usage
- Improve LSP performance and increase Python/TypeScript memory limits

Subscribe to Telegram updates:

Cursor v2.0 released Features - Run up to eight agents in parallel - Add Composer agentic coding model (4x faster) - Promote Browser to GA and embed in-editor - Make sandboxed terminals GA and default on macOS - Offer Cloud Agents with 99.9% reliability and instant startup - Add Team Commands for centrally managed team rules - Allow building plans in background with another model - Improve prompt UI: show files as inline pills and let Agent self-gather context Fixes - Fix memory leaks and reduce overall memory usage - Improve LSP performance and increase Python/TypeScript memory limits Subscribe to Telegram updates:

TLDR; I make software with AI and talk about it. Serial internet biz builder with multiple 6 & 7 figure exits. Always learning.

avatar for Ian Nuttall
Ian Nuttall
Wed Oct 29 18:55:58
A federal grand jury has indicted progressive Illinois House candidate Kat Abughazaleh for allegedly assaulting and conspiring to injure law enforcement during a protest last month at an ICE facility near Chicago.

A federal grand jury has indicted progressive Illinois House candidate Kat Abughazaleh for allegedly assaulting and conspiring to injure law enforcement during a protest last month at an ICE facility near Chicago.

The only official ABC News account. Download our mobile app for the latest updates: https://t.co/LgW7Q5IRpv

avatar for ABC News
ABC News
Wed Oct 29 18:55:04
Rio Governor Claudio Castro said following the deadliest police operation in Brazil's history that the raids 'delivered a hard blow to crime.' At least 132 people were killed in raids in Rio de Janeiro, officials said

Rio Governor Claudio Castro said following the deadliest police operation in Brazil's history that the raids 'delivered a hard blow to crime.' At least 132 people were killed in raids in Rio de Janeiro, officials said

Top and breaking news, pictures and videos from Reuters. For breaking business news, follow @ReutersBiz. Our daily podcast is here: https://t.co/KO0QFy0d3a

avatar for Reuters
Reuters
Wed Oct 29 18:55:00
Powell said Fed officials are struggling to reach a consensus about what's next for monetary policy and that financial markets should not assume another interest-rate cut will happen at the end of the year. Follow live updates:

Powell said Fed officials are struggling to reach a consensus about what's next for monetary policy and that financial markets should not assume another interest-rate cut will happen at the end of the year. Follow live updates:

Top and breaking news, pictures and videos from Reuters. For breaking business news, follow @ReutersBiz. Our daily podcast is here: https://t.co/KO0QFy0d3a

avatar for Reuters
Reuters
Wed Oct 29 18:53:51
WATCH LIVE: Sen. Grassley provides update on ‘Arctic Frost’ investigation

WATCH LIVE: Sen. Grassley provides update on ‘Arctic Frost’ investigation

Watch live coverage as Sen. Grassley provides an update on the ‘Arctic Frost’ investigation. Download here:

avatar for Fox News
Fox News
Wed Oct 29 18:53:28
this post is complete misinformation

LLMs are lossy compressors!  of *training data*.

LLMs losslessly compress *prompts*, internally. that’s what this paper shows.

source:  i am the author of “Language Model Inversion”,  the original paper on this

this post is complete misinformation LLMs are lossy compressors! of *training data*. LLMs losslessly compress *prompts*, internally. that’s what this paper shows. source: i am the author of “Language Model Inversion”, the original paper on this

phd research @cornell // language models, information theory, science of AI

avatar for Jack Morris
Jack Morris
Wed Oct 29 18:53:09
  • Previous
  • 1
  • More pages
  • 1922
  • 1923
  • 1924
  • More pages
  • 2131
  • Next