@TEG@mastodon.online
@TEG@mastodon.online avatar

TEG

@TEG@mastodon.online

Interested in psychology, AI, philosophy of science, methodology, religion (Matthew 23 /heart), other random things. #psychology #science #cognitive #ai

This profile is from a federated server and may be incomplete. Browse more on the original instance.

kellogh, to random
@kellogh@hachyderm.io avatar

this is in reference to super-alignment & safety, but my cousin also had her DEI team disbanded and “distributed” in the same way

on the surface, i think safety, DEI, and similar topics should be embedded in the culture and not centralized into a specific team. centralization would cause people to say, “oh that’s not my job”.

then again, any time a centralized team is disbanded, my immediate thought is, “apparently safety/DEI/etc. doesn’t matter to this company”. it’s a paradox, i suppose

TEG,
@TEG@mastodon.online avatar

@kellogh I always feel a little conflicted about reports like this. Like, it's 100% a good and important thing in general, but that doesn't mean a specific person or team or culture engaged with AI safety automatically inherits that value regardless of what they're actually contributing.

That said, I do think it might need a dedicated if small team to ensure that things are widely embedded.

TEG, to random
@TEG@mastodon.online avatar

Paywalled (and possibly blocked?) but looks very interesting... https://www.nytimes.com/2024/05/16/world/europe/new-yorker-story-murder-letby-britain.html?utm_source=pocket-newtab-en-gb

I don't know whether Letby is innocent or not of course but everything I happened to see about the evidence seemed very iffy and Lucia de B.-reminiscent.

TEG, to random
@TEG@mastodon.online avatar

Oh this was a new one for me - someone on a show did a "Hey Google" joke and it activated my phone.

bespacific, to generativeAI
@bespacific@newsie.social avatar

Fake studies have flooded publishers of top leading to thousands of , M of $ in lost revenue. Biggest hit has come to 217-year-old based in Hoboken NJ which announced it is closing 19 journals, some of which were infected by large-scale research . Wiley has reportedly had to retract more than 11,300 papers recently “that appeared compromised” as makes it easier for paper mills to peddle fake research. https://www.wsj.com/science/academic-studies-research-paper-mills-journals-publishing-f5a3d4bc

TEG,
@TEG@mastodon.online avatar

@bespacific I've always had a special hatred for a particular kind of review request/paper, where you can feel so clearly there's something wrong with it, but it's very hard to put your finger on what exactly. It's just slightly rotten everywhere. I found them hard to reject in a review becasue they're slippery in that way, "it's stinky" probably won't cut it.

Had one of those review requests a day or so ago and I strongly suspect AI was used from the Abstract, but how do you nail it down?

solalnathan, to academia
@solalnathan@sigmoid.social avatar

Isn't it weird that acceptance rate is a thing we look for in a conference/journal?

Publishing a paper should not be competitive like "we take the top 20% paper", it should be "we take all papers that are good enough according to our standards". Sometimes it can be a very low or very high number depending on the quality of the paper submitted.

@academicchatter @phdstudents

TEG, (edited )
@TEG@mastodon.online avatar

@solalnathan @academicchatter @phdstudents I wonder if it's all due to a lack of substantive understanding and expertise (or concern), at the level of decision-makers. "Top x% is excellent!", whether it's accepted papers or grants provided, is a totally substance-free metric. Any monkey can apply it and claim they're measuring exceptionality (just not what kind exactly).

To publish a paper if and only if it's "(a bit/very/extremely) valuable to a scientific field" requires much more judgment.

sideshow_jim, to mastodon
@sideshow_jim@mastodon.world avatar

Probably idiot question.
If I were to make a bot* in as a learning project, do I need a machine running constantly to run it, or should I do it on a cloud service?
I got a sitting around, would that do?

(*One that regularly posts a Josephine Baker photo, if you're interested)

TEG,
@TEG@mastodon.online avatar

@sideshow_jim I'm not an expert in the slightest, but FWIW I hacked together a Twitterbot (years ago) that I had running on PythonAnywhere.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • kavyap
  • DreamBathrooms
  • thenastyranch
  • magazineikmin
  • InstantRegret
  • GTA5RPClips
  • Youngstown
  • everett
  • slotface
  • rosin
  • osvaldo12
  • mdbf
  • ngwrru68w68
  • megavids
  • cubers
  • modclub
  • normalnudes
  • tester
  • khanakhh
  • Durango
  • ethstaker
  • tacticalgear
  • Leos
  • provamag3
  • anitta
  • cisconetworking
  • lostlight
  • All magazines
    •