e chi retweeted
Sydney Sweeney reveals in an interview that she doesn’t care about the PHP vulnerability because she moved everything to Rust in 2024.
“They said the rewrite would take six months. We’re on month 30. We haven’t shipped a feature since, but I’m having a very normal Wednesday.”
e chi retweeted
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fif…
This Post is from an account that no longer exists. Learn more
e chi retweeted
>arrive at HRT party
>ask the door girl if it's quant finance or estrogen
>she doesnt understand
>pull out illustrated diagram explaining the difference between quant finance and estrogen
>she laughs and says “it's a good party sir”
>go inside
>it's an estrogen party
e chi retweeted
"Do you think, my disciples, that I have any concealments? I conceal nothing from you. My disciples, there is nothing which I do that is not shown to you — that is my Way."
e chi retweeted
What is the role of human creativity in the autoresearch era? We imagine a world of abundant research where no idea is left untouched, and discovery becomes unbounded
h/t @chang_trenton @quadrillionlabs quadrillion.ai/blog/research…
e chi retweeted
ChatGPT came out midway through my PhD, and I graduated just in time to escape (or miss out on!) agent-driven research.
Autoresearch can boost discovery. But in its worst form, it'll hold us back. Research isn't results, and I fear we’ve Goodhart-ed ourselves into believing that.
Let's talk about how we steer research towards an abundance of discovery, not an abundance of papers.
quadrillion.ai/blog/research…
w/ @quadrillionlabs
"Imagine an earring that whispers helpful tips in your ear: “Better for you to set the learning rate to 0.00251342.” Good results are no longer in question; with a Whatever Agent Ultra Max subscription, you never come to a meeting short on quantitative results or new experimental ideas...
In the original story, the earring’s advice is always right, and the people who wear it live remarkably successful lives. But when they die, the part of their brain responsible for deliberation has withered from disuse.."
h/t @chang_trenton
quadrillion.ai/blog/research…
e chi retweeted
how did natural language processing researchers deal with the rise of natural language processing models and their application to almost every other domain?
… where do these people think the grads leading almost every frontier lab training team come from
e chi retweeted
example output from my initial looming with this model. trained using ~4 B200 hours from @nebiusai on 40M tokens of high quality greentexts sourced from a variety of subreddits, cleaned and OCR-ed with gpt 6 luna. imo this model achieves my initial research goal and shows an emergent ability to produce humor, and i expect to see results improve with scale as well as more diverse datasets
everyone cares about autonomous AI research (including @GoogleDeepMind)
blog.google/innovation-and-a…
e chi retweeted
New paper.
Previous: Models transfer a love of owls through sequences of numbers (Subliminal Learning).
New: We show models can transfer more complex traits: novel skills, agentic hacking, backdoors. Here, a backdoor transfers with neither trigger nor behavior in the data!
