Berna Devezer

A revised version of this preprint is now posted on arXiv. The theory section is rewritten, improving model exposition based on feedback we received, we fixed some confusion and errors, contextualized the models more carefully so the interpretation becomes clearer, and improved the overall writing.

VOR of this paper is finally live! muse.jhu.edu/pub/1/articl...

The world needs a high-quality, truly open language model. I wrote about why it’s necessary and why it’s easier to build than you might think.

little bit of a tangent: i once had a department chair who physically threatened me because the graduate student he was having an affair with was struggling in my class. i ended up resigning a few months later. so dating a student can have far-reaching consequences beyond the abuse of the student.

"AI this, you m'fers!"

idea in the shower --> prompt --> chatbot --> paper --> chatbot --> reviewers --> chatbot --> reubuttal --> chatbot --> associate editor --> chatbot --> random number generator --> acceptance --> CV --> profit

hear hear

The purpose of peer review is not primarily to help authors improve their papers.

Chris Donkin and I make a similar point here: doi.org/10.1080/0951... Ironically, It's a concept that invites misuse. Accusations of "QRP" are easy to make and difficult to dispute, even with preregistration etc.

My experience has been that some of these folks (not all, but some) they believe their *own* practices are unquestionable, so there is that

a few of these guys have me blocked from 2020 on twitter — some have called me names because of this retraction we helped facilitate (see right side "Responses" www.eneuro.org/content/6/6/...) — so I have no idea what they are up to here nor on twitter (because I also do not go there at all), but...

Anyone in academia (or anywhere else) know anything about Korn Ferry Consulting? They are going to run the search for the new president at the University of Idaho. Would greatly appreciate any info. #highered #academia #academicsky www.kornferry.com

A judge on Wednesday ordered Meta to pay New Mexico $25 billion after a jury found the company liable for deceiving users about privacy and content moderation policies on Facebook.

ad copy is probs one of the worst places to use llm-generated text. both bec it's so sparse that there's nothing to hide behind and it's the last place you want to sound generic or cliché. just watched an online ad for vuori that started with "every performance has a before and an after." o rly 😂

Enjoyed seeing this woman's work. Sustainability isn't just about producer decisions, but also consumer decisions, namely their willingness to repair things when needed. This type of repair is easier on natural fibers, rather than stretch fabrics. Consider that when shoppping. IG loosethreadsrepair

Some notable examples of long-standing errors in science from Douglas Allchin’s new article: “When is science ‘self-correcting’?”: doi.org/10.1016/j.sh... Few quotes follow…

is this a description of being middle aged?

Helen’s obituary in the New York Times will start to give you a sense of just how remarkable she was as a reporter and public health expert.

Fast on the heels of finally settling on a title, Yale just sent my new book cover, and I love it very much.

finally made my double-vaxx appointments for friday. fingers crossed i won't catch anything until then. thanks everyone who's been posting about getting theirs or i'd have forgotten about it.

I will be looking for a PhD student in Bayesian statistics (in collaboration with Paul Bürkner), located in Prague. Forward to students if you have somebody who could be interested. www.martinmodrak.cz/grammo-call/

If you're backyard is close to water, consider building a frog sauna to help treat chytrid!

I see the same thing regularly on here with people complaining about economics. It's a brainworm that could be cured just by taking a couple of higher level economics courses. Always easier to find and crucify an out group rather than accept that things are nuanced.