Mark Norris 🇺🇦 retweeted
1) As @WVU_Worldlangfaculty, students’ right to quality education is always our priority, so prior to our appeal on 8/25, we sent a survey to 2,640 students who are currently enrolled in our UG classes. 1,755 answered. Full report point: worldlanguages.wvu.edu/about 👇thread
Mark Norris 🇺🇦 retweeted
That is, people have tried refutation in print, they've tried addressing the "merits" of arguments, they've tried pointing out the harms and the bad faith of it all and still nothing changes. So the only recourse left is literally deplatforming bigots.
Mark Norris 🇺🇦 retweeted
It's been 20 years, time for a new DUDE survey. Please share widely!
pitt.co1.qualtrics.com/jfe/f…
Mark Norris 🇺🇦 retweeted
I don't know if I'll ever truly organize my thoughts on this, but let's also normalize and destigmatize non-industry jobs post-academia.
Like, I'm so much happier not sitting in front of a computer or at a desk.
Mark Norris 🇺🇦 retweeted
GPTchat gets enough stuff fundamentally wrong that an assignment like this would work really well
This quoted post is unavailable.
Mark Norris 🇺🇦 retweeted
I like when OCR randomly capitalizes letters in PDF metadata, making it sound like the National Endowment for the Arts is being really sarcastic
Mark Norris 🇺🇦 retweeted
Language is one important tool we have for communicating with other people, sharing ideas, & coordinating our actions. But just because language is a medium we use to become aware of the internal lives of other people doesn't mean the language itself is the internal life.
>>
Mark Norris 🇺🇦 retweeted
The sequences of words are just externally visible artifacts that we use in the extended communicative acts that are a core part of education---and that others can observe as well.
Mark Norris 🇺🇦 retweeted
I'm looking for a postdoc in syntax, just approved for the upcoming year! Let me know if you know someone appropriate & potentially interested
Mark Norris 🇺🇦 retweeted
Seriously: the claim that one has to understand the internals of an LLM to be able to evaluate claims about what it can and can't do is a prime example of what you so aptly called "hierarchies of knowledge", @timnitGebru
Comment on @nytimes article saying that ppl like @emilymbender "don't know the math" & that's why they're criticizing.
Is it an Open AI VP who understands the ins & outs & "math" of LLMs rather than someone whose been studying language for decades?
Mark Norris 🇺🇦 retweeted
#sociolinguistics Twitter, help (please RT)!
I'm looking for a .pdf of the Labov paper where he takes a rabbit (at first hidden under a trenchcoat lol) into a classroom, then tells the kids the bunny will get nervous unless they talk to it, then leaves (1/n)
Mark Norris 🇺🇦 retweeted
I refuse to be delegated to the "skeptics box" in someone else's framing of a debate. Here is my response to @stevenbjohnson 's NYT Magazine article about LLMs and OpenAI.
On NYT Magazine on AI: Resist the Urge to be Impressed link.medium.com/rag8kbGujpb
Mark Norris 🇺🇦 retweeted
The job market is so bad there hasn’t been any new Affistant Profeffor positions since 1800
Mark Norris 🇺🇦 retweeted
This paper claims to follow best practice by including a datasheet and a model card, but really all it does is reinforce that 780B tokens is WAY too big to be sufficiently documented. >>
storage.googleapis.com/pathw…
Mark Norris 🇺🇦 retweeted
Sad news: Lauri Karttunen passed away peacefully this morning. Lauri was a towering figure in linguistics and NLP, and a vibrant presence at Stanford in Linguistics, @stanfordnlp & @StanfordCSLI. So many observations and concepts that we all take for granted trace to his work!
Mark Norris 🇺🇦 retweeted
incredibly evil
Mark Norris 🇺🇦 retweeted
As we #StandWithUkraine, we are updating our product with a message that provides resources for helping the people of Ukraine. We have also suspended our service in Russia and Belarus.
Learn more: gram.ly/WeStandWithUkraine
























