Jorge Hernandez 🇺🇦 🏳️🌈
@braneloop.bsky.social
Principal ML Engineer • AuDHD • Tweets: ML/AI, Math, Neuroscience, Physics, Philosophy, other stuff.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
The Oculus is the main station house of the Port Authority Trans-Hudson (PATH) rapid Transit system which connects Manhattan to New Jersey. It opened on March 3, 2016, as part of the reconstruction of the World Trade Center complex after the 2001 terrorist attacks.
#Photography #NewYork #USA
#Photography #NewYork #USA
October 1, 2025 at 12:18 AM
The Oculus is the main station house of the Port Authority Trans-Hudson (PATH) rapid Transit system which connects Manhattan to New Jersey. It opened on March 3, 2016, as part of the reconstruction of the World Trade Center complex after the 2001 terrorist attacks.
#Photography #NewYork #USA
#Photography #NewYork #USA
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Chronic poor sleep quality, inflammation, and accelerating brain aging (BAG=brain age gap)
www.thelancet.com/journals/ebi...
www.thelancet.com/journals/ebi...
October 1, 2025 at 4:29 PM
Chronic poor sleep quality, inflammation, and accelerating brain aging (BAG=brain age gap)
www.thelancet.com/journals/ebi...
www.thelancet.com/journals/ebi...
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
My conversation today with @drtomfrieden.bsky.social, Former CDC Director, on his new book THE FORMULA and our current state of affairs in US public health
erictopol.substack.com/p/tom-friede...
erictopol.substack.com/p/tom-friede...
October 1, 2025 at 9:02 PM
My conversation today with @drtomfrieden.bsky.social, Former CDC Director, on his new book THE FORMULA and our current state of affairs in US public health
erictopol.substack.com/p/tom-friede...
erictopol.substack.com/p/tom-friede...
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
I'm hiring for a PhD position at Malmö University, Sweden!
The project will focus on molecular modelling of proteins, lipids, and biomolecular condensates at cell membranes.
More details and application form: tinyurl.com/4zm92365
Please feel free to share!
@vetenskapsradet.bsky.social | @mau.se
The project will focus on molecular modelling of proteins, lipids, and biomolecular condensates at cell membranes.
More details and application form: tinyurl.com/4zm92365
Please feel free to share!
@vetenskapsradet.bsky.social | @mau.se
October 2, 2025 at 7:46 AM
I'm hiring for a PhD position at Malmö University, Sweden!
The project will focus on molecular modelling of proteins, lipids, and biomolecular condensates at cell membranes.
More details and application form: tinyurl.com/4zm92365
Please feel free to share!
@vetenskapsradet.bsky.social | @mau.se
The project will focus on molecular modelling of proteins, lipids, and biomolecular condensates at cell membranes.
More details and application form: tinyurl.com/4zm92365
Please feel free to share!
@vetenskapsradet.bsky.social | @mau.se
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Test the activity of 300+ natural enzymes against 100+ substrates, discover 200+ new enzymatic reactions, and train machine learning models to predict which enzymes can do which reactions.
@aepaton.bsky.social @gabegomes.bsky.social @alisonnarayan.bsky.social
www.nature.com/articles/s41...
@aepaton.bsky.social @gabegomes.bsky.social @alisonnarayan.bsky.social
www.nature.com/articles/s41...
October 2, 2025 at 8:11 PM
Test the activity of 300+ natural enzymes against 100+ substrates, discover 200+ new enzymatic reactions, and train machine learning models to predict which enzymes can do which reactions.
@aepaton.bsky.social @gabegomes.bsky.social @alisonnarayan.bsky.social
www.nature.com/articles/s41...
@aepaton.bsky.social @gabegomes.bsky.social @alisonnarayan.bsky.social
www.nature.com/articles/s41...
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
What the generals, admirals and their senior enlisted face when they return from Quantico. My latest in @thebulwark.com
www.thebulwark.com/p/questions-...
www.thebulwark.com/p/questions-...
Questions After Quantico
U.S. military commanders around the world will be working to provide steady leadership after Hegseth’s bravado and Trump’s bluster.
www.thebulwark.com
October 3, 2025 at 12:28 PM
What the generals, admirals and their senior enlisted face when they return from Quantico. My latest in @thebulwark.com
www.thebulwark.com/p/questions-...
www.thebulwark.com/p/questions-...
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
The opportunities and risks of the entry of LLMs into mathematical research in one screenshot. I think it is clear that LLMs will make trained researchers more effective. But they will also lead to a flood of bad/wrong papers, and I'm not sure we have the tools to deal with this.
September 24, 2025 at 8:00 PM
The opportunities and risks of the entry of LLMs into mathematical research in one screenshot. I think it is clear that LLMs will make trained researchers more effective. But they will also lead to a flood of bad/wrong papers, and I'm not sure we have the tools to deal with this.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
working on a new unified theory of american reality i'm calling "everyone is twelve now"
September 8, 2025 at 12:37 AM
working on a new unified theory of american reality i'm calling "everyone is twelve now"
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Adi Shamir's advice to young researchers:
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
March 26, 2025 at 11:24 AM
Adi Shamir's advice to young researchers:
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
1. Read, read, read. Back in the eighties, I read every cryptography paper out there. Once that became impossible, I read the abstract of every paper. Now I read at least every title.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
one of the wildest bits of discourse, is the idea that we can just ban ai
we cannot ban ai. code is speech and you can run these models on a 4080. there's no putting this genie back in the bottle.
we cannot ban ai. code is speech and you can run these models on a 4080. there's no putting this genie back in the bottle.
August 27, 2025 at 4:10 AM
one of the wildest bits of discourse, is the idea that we can just ban ai
we cannot ban ai. code is speech and you can run these models on a 4080. there's no putting this genie back in the bottle.
we cannot ban ai. code is speech and you can run these models on a 4080. there's no putting this genie back in the bottle.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
I've gotta comment on this:
A lot of people on BlueSkyhere think this is evidence that generative AI isn't useful, but that's not necessarily the correct conclusion, IMO.
What this may tell us is that generative AI is not something you can just add onto existing company systems as is.
#MLSky
A lot of people on BlueSkyhere think this is evidence that generative AI isn't useful, but that's not necessarily the correct conclusion, IMO.
What this may tell us is that generative AI is not something you can just add onto existing company systems as is.
#MLSky
An M.I.T. study found that 95% of companies that had invested in A.I. tools were seeing zero return. It jibes with the emerging idea that generative A.I., “in its current incarnation, simply isn’t all it’s been cracked up to be,” johncassidysays.bsky.social writes.
The A.I.-Profits Drought and the Lessons of History
Like the steam engine, electricity, and computers, generative artificial intelligence could take longer than expected to transform the economy.
www.newyorker.com
August 25, 2025 at 9:02 PM
I've gotta comment on this:
A lot of people on BlueSkyhere think this is evidence that generative AI isn't useful, but that's not necessarily the correct conclusion, IMO.
What this may tell us is that generative AI is not something you can just add onto existing company systems as is.
#MLSky
A lot of people on BlueSkyhere think this is evidence that generative AI isn't useful, but that's not necessarily the correct conclusion, IMO.
What this may tell us is that generative AI is not something you can just add onto existing company systems as is.
#MLSky
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
When reading AI reasoning text (aka CoT), we (humans) form a narrative about the underlying computation process, which we take as a transparent explanation of model behavior. But what if our narratives are wrong? We measure that and find it usually is.
Now on arXiv: arxiv.org/abs/2508.16599
Now on arXiv: arxiv.org/abs/2508.16599
Humans Perceive Wrong Narratives from AI Reasoning Texts
A new generation of AI models generates step-by-step reasoning text before producing an answer. This text appears to offer a human-readable window into their computation process, and is increasingly r...
arxiv.org
August 27, 2025 at 9:30 PM
When reading AI reasoning text (aka CoT), we (humans) form a narrative about the underlying computation process, which we take as a transparent explanation of model behavior. But what if our narratives are wrong? We measure that and find it usually is.
Now on arXiv: arxiv.org/abs/2508.16599
Now on arXiv: arxiv.org/abs/2508.16599
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Excellent talk yesterday at UTS by Seyoon Ragavan, on #quantum algorithms for factoring: focusing on Regev's recent improvement, and subsequent improvements by Seyoon and his coauthors.
August 27, 2025 at 9:58 PM
Excellent talk yesterday at UTS by Seyoon Ragavan, on #quantum algorithms for factoring: focusing on Regev's recent improvement, and subsequent improvements by Seyoon and his coauthors.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
This is obviously a fantastic initiative, which everyone who wishes to enter the US will want to applaud enthusiastically.
June 26, 2025 at 6:06 PM
This is obviously a fantastic initiative, which everyone who wishes to enter the US will want to applaud enthusiastically.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
As far as I can tell, HoTT is about making types into (∞,0)-categories (aka ∞-groupoids), and the directed HoTT folks want to make types into (∞,∞)-categories (aka ∞-categories), but I reckon you could already get a lot of mileage if you could make types into (∞,1)-categories (aka categories, up […]
Original post on mathstodon.xyz
mathstodon.xyz
June 26, 2025 at 6:55 PM
As far as I can tell, HoTT is about making types into (∞,0)-categories (aka ∞-groupoids), and the directed HoTT folks want to make types into (∞,∞)-categories (aka ∞-categories), but I reckon you could already get a lot of mileage if you could make types into (∞,1)-categories (aka categories, up […]
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
I just read the heck out of this long blog post and now I know who invented backprop
https://yuxi-liu-wired.github.io/essays/posts/backstory-of-backpropagation/
https://yuxi-liu-wired.github.io/essays/posts/backstory-of-backpropagation/
The Backstory of Backpropagation – Yuxi on the Wired
Why backprop was resisted for 20 years: assumption of discretely spiking neurons, goal of synthesizing Boolean logic, fear of local optima, and bad luck. Werbos has the best claim for invention.
yuxi-liu-wired.github.io
May 25, 2025 at 7:04 AM
I just read the heck out of this long blog post and now I know who invented backprop
https://yuxi-liu-wired.github.io/essays/posts/backstory-of-backpropagation/
https://yuxi-liu-wired.github.io/essays/posts/backstory-of-backpropagation/
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Flash invariant point attention enables the generation of larger biomolecules.
arxiv.org/abs/2505.11580
arxiv.org/abs/2505.11580
May 27, 2025 at 5:11 PM
Flash invariant point attention enables the generation of larger biomolecules.
arxiv.org/abs/2505.11580
arxiv.org/abs/2505.11580
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
For what it's worth, NEJM, JAMA and the Lancet are basically the medical equivalent of, like, Science, Nature and Cell.
HHS Secretary Robert F. Kennedy Jr. threatened to bar government scientists from publishing in major medical journals, saying on a podcast that three top journals were “corrupt” and influenced by pharmaceutical companies.
RFK Jr. threatens to bar government scientists from publishing in leading medical journals
The health secretary said the New England Journal of Medicine, the Journal of the American Medical Association, and The Lancet are in bed with pharma.
www.politico.com
May 27, 2025 at 8:32 PM
For what it's worth, NEJM, JAMA and the Lancet are basically the medical equivalent of, like, Science, Nature and Cell.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
1. It's Fustilarian Friday and today's drop is the "Restoring Gold Standard Science" executive order.
My read is that this is designed to (1) protect far-fringe figures such as the antvaxxers now heading our health agencies, and (2) kneecap efforts to use scientific evidence in regulatory policy.
My read is that this is designed to (1) protect far-fringe figures such as the antvaxxers now heading our health agencies, and (2) kneecap efforts to use scientific evidence in regulatory policy.
Restoring Gold Standard Science
By the authority vested in me as President by the Constitution and the laws of the United States of America, including section 7301 of title 5, United
www.whitehouse.gov
May 24, 2025 at 5:24 AM
1. It's Fustilarian Friday and today's drop is the "Restoring Gold Standard Science" executive order.
My read is that this is designed to (1) protect far-fringe figures such as the antvaxxers now heading our health agencies, and (2) kneecap efforts to use scientific evidence in regulatory policy.
My read is that this is designed to (1) protect far-fringe figures such as the antvaxxers now heading our health agencies, and (2) kneecap efforts to use scientific evidence in regulatory policy.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
The first rule of the One-Way Communication Complexity club is that Bob doesn't talk about the One-Way Communication Complexity club
May 27, 2025 at 8:49 PM
The first rule of the One-Way Communication Complexity club is that Bob doesn't talk about the One-Way Communication Complexity club
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
I wish more people understood the difference between criticizing something because you want to fix it and criticizing something because you want to destroy it.
May 25, 2025 at 1:51 PM
I wish more people understood the difference between criticizing something because you want to fix it and criticizing something because you want to destroy it.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
Titanic& super important work from @drmichaellevin Lab
"What Lives? A meta-analysis of diverse opinions on the definition of life
I’m on the dynamic and relational process side
close to self-sustaining dynamic patterns 😎👇🏼
arxiv.org/abs/2505.15849
osf.io/preprints/ps...
"What Lives? A meta-analysis of diverse opinions on the definition of life
I’m on the dynamic and relational process side
close to self-sustaining dynamic patterns 😎👇🏼
arxiv.org/abs/2505.15849
osf.io/preprints/ps...
May 26, 2025 at 6:30 AM
Titanic& super important work from @drmichaellevin Lab
"What Lives? A meta-analysis of diverse opinions on the definition of life
I’m on the dynamic and relational process side
close to self-sustaining dynamic patterns 😎👇🏼
arxiv.org/abs/2505.15849
osf.io/preprints/ps...
"What Lives? A meta-analysis of diverse opinions on the definition of life
I’m on the dynamic and relational process side
close to self-sustaining dynamic patterns 😎👇🏼
arxiv.org/abs/2505.15849
osf.io/preprints/ps...
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
I did not attend QIP 2025 partially because I did not want to risk crossing the US border.
QIP should not be in the US until it is safe for everyone to travel there.
QIP should not be in the US until it is safe for everyone to travel there.
It pains me to say this, but if you are not a US citizen, you should not attend scientific conferences in the US, visit US institutions, or otherwise travel to our country.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.
May 26, 2025 at 6:15 PM
I did not attend QIP 2025 partially because I did not want to risk crossing the US border.
QIP should not be in the US until it is safe for everyone to travel there.
QIP should not be in the US until it is safe for everyone to travel there.
Reposted by Jorge Hernandez 🇺🇦 🏳️🌈
It pains me to say this, but if you are not a US citizen, you should not attend scientific conferences in the US, visit US institutions, or otherwise travel to our country.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.
May 26, 2025 at 1:54 AM
It pains me to say this, but if you are not a US citizen, you should not attend scientific conferences in the US, visit US institutions, or otherwise travel to our country.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.
Right now it's not worth the risk.
Hopefully that will change. If it doesn't, science in the US is sunk anyway.