mt-b

I too am very curious about the Chinese sources! Sounds like a great project though!!

ngl I was taught a lot of this as a middle schooler in Michigan but I kinda assumed it was mostly local propaganda

gonna be harder than ever to get that story out now, IMO

source: 《职位分类与人事管理》陈子明、李盛平 （1986）

I will never get tired of these choose your fighter-lookin-ass figures

i see "disjuncture" and raise you "discontinuity"

Announcing the release of new Common Corpus, and the first really global update. For the Indian AI Impact Summit we added 270B tokens of permissibly license collections, more than half of content from non-Western countries. pleias.fr/blog/blogcom...

Did Peng Zhen Rebel Against Mao?

A startling Chinese-language biography of Mao’s one-time enforcer reveals the folly of hoping that Party officials will break ranks — and the importance of revisionist histories.

Read Joseph Torigian's review-essay: chinabooksreview.com/2026/02/19/p...

obsequious thin-generalists with a main talent of making slides and recommending the same seven things in every case can never be replaced by ai, kevin

lobsters fighting meme, both lobsters and the cheering people all labeled "STATISTICIAN"

these memes are fresh

This is where my center of gravity lives too. Thanks!

polysemy strikes again! In context, people in my field normally mean “I embedded the documents [in a vector space]” when they say “we embedded the documents” rather than that kind of embedding!

bumping for the evening crowd bc I am still curious about this

right, yes. just like weirdly rare to hear people specify in eg presentations ? thanks for weighing in, I feel less bad for assuming now

returning to this question while I stew about document embeddings architectures ( just for fun )

thanks Paul 🙏

yeah I mean 1 is the big concern. but also: I haven’t hung with methods or NLP people in a while and I don’t want to be an ass just to reveal ~no information to anyone other than myself

for me, I tend to assume that this means “we averaged the word embeddings“ unless someone specifies otherwise.

I’m trying to figure out if I have missed convergence on this or if I should interrupt people to say “and how did you do that”

ok team. what “embedding” procedure do we expect researchers have followed when they say “we embedded the documents and compared them”?

comment below ⬇️

Horse year is galloping in.
I hope yours is a bewty…
#China #Beijing

Which is why I could write something like this before this technology could even find its feet. We're talking about desires that predate this particular technology, and, in fact, have always attached themselves to computers.
bsky.app/profile/kevi...

GitHub - andrewheiss/null-world-testing Contribute to andrewheiss/null-world-testing development by creating an account on GitHub.

oh, all the Quarto/R/OJS code is here too: borrow/adapt away! github.com/andrewheiss/...

surprising no one, I have read only 1 book and 1 article and written basically nothing

good news. I have narrowed it down to only 4 books (and 5 articles) I need to read today before I "can" write

kieranhealy.org/blog/archive...

Automate the C-Suite - The Ideas Letter Weatherby argues that Benanav’s and Morozov’s essays on AI and the left miss a deeper reality: Today’s AI is not a neutral tool operating within capitalism but the materialization of…

www.theideasletter.org/essay/automa...
h/t @laurenfklein.bsky.social

I have been working on something to try to share more materials in a more systematic way but it requires confronting both some risk and the shameful disorganization that characterizes my pile of hoarded material

data hoarders unite

It was about the University of the Andes, I think?

Visualizing China's leadership: I'm back with an updated infographic, this time showing newly purged party members, new roles, and updated job descriptions.

Download full version: jagevanscom.wordpress.com/wp-content/u...