steifer.pl - Profile | ThreadSky | a Reddit-style client for Bluesky

theonion.com/this-war-wil...

submitted 13 days ago • 119 comments

I can't say I understand why this inkwell from the 16th century was made, but it's well done.

submitted 35 days ago • 3 comments

Yesterday, the editorial board of Mathematical Logic Quarterly resigned and announced creation of a new diamond open access journal: Zeitschrift für Mathematische Logik und Grundlagen der Mathematik. Let's say I am just going to call it ZML.

submitted 83 days ago • 0 comments

AHAHAHAHAHAHAHAHA WELCOME TO HELL, AI INDUSTRY

submitted 89 days ago • 5 comments

The disconnect between AI benchmarks and math research (Evaluating AI systems on their ability to be a mathematical copilot). ~ Ralph Furman. sugaku.net/content/ai-b... #AI #LLMs #Math

submitted 90 days ago • 0 comments

Victor Cringeban.

submitted 91 days ago • 0 comments

A common remark from AI/ML conference reviewers is: "The paper lacks a clear message." This attitude goes opposite to how research is done. Sometimes, the message is the entire paper and it can't be compressed into 140 signs. Simplified messages mislead and are usually false. #academicsky #MLsky

submitted 92 days ago • 0 comments

A common remark from AI/ML conference reviewers is: "The paper lacks a clear message." This attitude goes opposite to how research is done. Sometimes, the message is the entire paper and it can't be compressed into 140 signs. Simplified messages mislead and are usually false. #academicsky #MLsky

submitted 92 days ago • 0 comments

Possibly the dumbest of the many dumb plagiarism claims nationalpost.com/news/mark-ca...

submitted 93 days ago • 17 comments

Yet another LLM-generated review. I knew something was off when they said our results were 'groundbreaking'. 🤨

submitted 93 days ago • 1 comment

This might be the best one yet!

submitted 96 days ago • 0 comments

Bluesky's gonna really take off once people discover joking

submitted 104 days ago • 140 comments

#academicsky

submitted 105 days ago • 0 comments

I applied to a grant scheme for establishing a research team and lost due to a lack of experience in leading research teams. The scheme is called 'First Team'. 😂

submitted 105 days ago • 0 comments

'My body is a temple' but it's Spain, 1936.

submitted 108 days ago • 0 comments

Cheers from AAAI-25 in Philadelphia! Any of my mutuals here?

submitted 123 days ago • 0 comments

A huge "DO NOT INGEST" written on my wife's shampoo bottle, but no one will tell me how to live my life, glurp glurp glurp glurp.

submitted 126 days ago • 1 comment

A huge "DO NOT INGEST" written on my wife's shampoo bottle, but no one will tell me how to live my life, glurp glurp glurp glurp.

submitted 126 days ago • 1 comment

it is very much not a delusion to assume infinite free images and text, because i can run something very close to state of the art on my desktop for everything other than video.

submitted 126 days ago • 6 comments

Replacing "I hope this email finds you well" with "Failure to respond will be taken as a resignation." in all emails from now on.

submitted 126 days ago • 3 comments

Yes.

submitted 128 days ago • 0 comments

New paper: Simulating Time With Square-Root Space people.csail.mit.edu/rrw/time-vs-... It's still hard for me to believe it myself, but I seem to have shown that TIME[t] is contained in SPACE[sqrt{t log t}]. To appear in STOC. Comments are very welcome!

submitted 128 days ago • 16 comments

This indeed sounds very cool.

submitted 131 days ago • 0 comments

I know in the humanities you can just use "positivist" to mean "silly wrong opinion", whatever, but it does kinda disappoint me to see this sorta thing in a history of science journal (www.journals.uchicago.edu/doi/full/10....). That.... that's not what positivists say about measurement conventions.

submitted 131 days ago • 5 comments

I am not convinced by this idea that AI tools are enabling any genocidal tendencies that you could not have found in your history textbook.

submitted 131 days ago • 0 comments

Great opportunity for central/east European philosophers of science.

submitted 136 days ago • 0 comments

Something for all you blue sky maniacs.

submitted 138 days ago • 0 comments

Last night I found out that the NSF math postdoctoral fellowship I applied for is being deleted because it does not comply with Trump’s executive orders on DEI in the federal government. I’m going to answer some FAQs and share some thoughts about this ordeal in this thread 1/n

submitted 142 days ago • 51 comments

This is (I believe) the first time cultivated meat has been available for sale in Europe. In a WORLD first, it's available as pet food! When I started talking and writing about this a decade ago, people thought the idea was laughable. www.theguardian.com/environment/...

submitted 142 days ago • 1 comment

'Meta admitted to torrenting a controversial large dataset known as LibGen'. Impressive! I have been recently struggling to get even a single book from this 'controversial dataset'. I suppose it was just Meta eating up all the bandwidth.

submitted 142 days ago • 0 comments

This new draft claims to give lower bounds on chain-of-though iterations, with hard attention, needed to solve PARITY and some other tasks (which is pretty cool!) arxiv.org/abs/2502.02393

submitted 144 days ago • 0 comments

I can finally share our draft about: a) the first method for proving lower bounds against one-layer Transformers with infinite precision and b) a new scalable attention mechanism for tracking higher-order interactions between tokens.... (more info in the 🧵) #MLsky #ML #AI #CompSky 🧪

submitted 147 days ago • 1 comment