Javier Burroni
@jburroni.bsky.social
430 followers 160 following 150 posts
“Machines take me by surprise with great frequency” A. M. T.
Posts Media Videos Starter Packs
And highly regulated.
That is why I was so surprise.
Related: the first time I heard of companies actually using AI to replace people was in the insurance industry, specifically with some underwriting stuff. I was honestly quite surprised.
will be a zoom link? It looks very interestin!
Before covid, I talked a lot with Zenna Tavares about this topic. What you described from the paper might be the right approach to think about it. I’ll read the paper soonish.
Muy bueno todo, pero que gran vino el Saint Felicien
looks very interesting. Thank you for sharing!
Yeah, I got that. thanks for sharing all the papers!
When I write (latex or code), I use vscode with multiple splits for the same reason
In my iPad I use an app named marginnotes that allows you to split the document in several windows and one is devoted to the appendix/references
Recently, I was thinking of going back to very low-level programming, but it wasn’t clear to me whether there was a real need for it or if it was just an echo-chamber effect. Your comment makes me think it’s the former.
Recuerdo haber hecho un comentario de esa índole (mutatis mutandis) sobre la serie “el puntero“
I ran the following experiment: for each person in the errata, I asked gpt to search for other books in which they appear. I learned about "Bayesian Filtering and Smoothing". Ended up buying it.
I'm very impressed by the complexity of the query and the difference in performance –Claude found nothing
looking forward to read the follow-up threads on this story
Reposted by Javier Burroni
Far less focus on consumer product feature, and much more on its development as an enabling technology for industrial automation and robotics. Take a snapshot of one of their plans, for instance. For them integrating AI into industrial production, and building [+]
Reposted by Javier Burroni
This is not OK.

I don't submit often to NeurIPS, but I reviewed papers for this conference almost every year. As a reviewer, why would I spend time trying to give a fair opinion on papers if it's what happens in the end???
This is the metareview
Reposted by Javier Burroni
Submit your work on to the @euripsconf.bsky.social workshop on on Differentiable Systems and Scientific Machine Learning!

differentiable-systems.github.io/workshop-eur...

Submission: 10 October
Notification: 31 October
Workshop: 6 or 7 December in Copenhagen
Differentiable Systems and Scientific Machine Learning Workshop - EurIPS 2025
differentiable-systems.github.io
Reposted by Javier Burroni
Unfortunately, our submission to #NeurIPS didn’t go through with (5,4,4,3). But because I think it’s an excellent paper, I decided to share it anyway.

We show how to efficiently apply Bayesian learning in VLMs, improve calibration, and do active learning. Cool stuff!

📝 arxiv.org/abs/2412.06014
Post-hoc Probabilistic Vision-Language Models
Vision-language models (VLMs), such as CLIP and SigLIP, have found remarkable success in classification, retrieval, and generative tasks. For this, VLMs deterministically map images and text descripti...
arxiv.org
Interesting feeling: I'm reading a book and, on the errata page, I see many acquaintances' names. Feels like I'm late to the party. (Clearly one I should’ve read earlier.)
Almost every ML post on tw feels like an ad in one way or another, producing different levels of cringe.
I wish I could see traces of principia Discordia online again.
Unrelated, but the permanent exposition of Sol LeWitt at MassMoCa is amazing.
Is this plot in real terms or nominal? (I assume the latter)
uv is the most important thing that happened to python in the last year
Are you using invisible color for the garlic?