Sign in

Fabio E. Tonti

@tontief.bsky.social
325 followers 655 following 237 posts

Absentee mathematician gone teaching. Aspiring statistician. Original AndOrNot_robot. Stats & maths education.

PostsRepliesMedia
Reposted by Fabio E. Tonti
Thomas Lumley @tslumley.bsky.social · 6h
A regression without colinearity diagnostics is like a fish without a bicycle
0213
Reposted by Fabio E. Tonti
Julian Togelius @togelius.bsky.social · 10/10/2026
New blog post: A.I.: Anti-Intellectualism togelius.blogspot.com/2026/10/ai-a...
311924
Fabio E. Tonti @tontief.bsky.social · 07/10/2026
I'm slowly starting to mute the LLMs-are-better-than-us-at-everything nihilist posters on here, so I can focus on interesting (and yes, reasonable) takes.
120
Fabio E. Tonti @tontief.bsky.social · 06/10/2026
Classic example of massive confusion in elites in recent decades. The elites are never in danger of being driven into suicide by "motivated" judicial and prosecution projects. We need to keep remembering him!
050
Reposted by Fabio E. Tonti
Emi Tanaka @emitanaka.org · 30/09/2026
I’m pleased to announce that Vol 18 Issue 3 of The R Journal is out now! 🥳 Free & open access, R journal is made possible by our volunteer editors, reviewers + contributing authors. Thank you all! journal.r-project.org/issues/2026-3/ 🧵 20 articles 👇 #rstats journal.r-project.org/issues/2026-3/
journal.r-project.org
The R Journal: Volume 18/3
Articles published in the September 2026 issue
13911
Reposted by Fabio E. Tonti
Darren Dahly @statsepi.bsky.social · 24/09/2026
Congrats to @jdwilko.bsky.social and everyone who contributed. www.bmj.com/content/394/...
bmj.com
INSPECT-SR tool for assessing trustworthiness of randomised controlled trials
The integrity of evidence synthesis is threatened by problematic randomised controlled trials, where there are serious concerns about the trustworthiness of the data or findings. Such concerns could b...
0176
Reposted by Fabio E. Tonti
Gary Hoppenworth @garytho.bsky.social · 22/09/2026
It turns out someone posted a third polyhedron with this property to arxiv a week ago lol arxiv.org/pdf/2609.177...
25312
Reposted by Fabio E. Tonti
Shubhendu Trivedi @shubhendu.bsky.social · 18/09/2026
These accounts exist aplenty. You could think of the entire Bourbaki project as concerned with understanding at a structural level, not problems. What is the right conceptual framework in which this theorem becomes natural? The emphasis was on definitions and conceptual architecture.
191
Fabio E. Tonti @tontief.bsky.social · 18/09/2026
I'm starting a new position as an (internal) stats consultant in an hta setting in November. Dear mutuals, I would really appreciate some of your favourite books/articles/ressources suggestions, especially on the (group) consulting topic:) @statsepi.bsky.social @ostlundollie.bsky.social
4111
Reposted by Fabio E. Tonti
Semiosmith @semiosmith.bsky.social · 13/09/2026
AI is good for giving the appearance of being productive, no?
011
Fabio E. Tonti @tontief.bsky.social · 10/09/2026
Haha well put!
000
Reposted by Fabio E. Tonti
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 04/09/2026
It's a systemic problem in science reporting that these summaries exaggerate the evidence. In the actual candy study, the participants were all students, status was manipulated in the lab by telling some students they were better than others, and "nearly twice" means 1 piece vs 2 pieces.
Piff and his colleagues found that upper-class drivers were more likely to cut other cars off at junctions and ignore pedestrians at crossings. Wealthier people also appeared more willing to lie at work. In one experiment, volunteers were left alone with a jar of sweets ostensibly meant for children. When the researchers returned, upper-class people had snaffled nearly twice as many candies as lower-class people.
1329166
Reposted by Fabio E. Tonti
Ryan Moulton @moultano.bsky.social · 02/09/2026
I think you all should know this bird exists.
lilac breasted roller
817116
Reposted by Fabio E. Tonti
Ben Recht @beenwrekt.bsky.social · 30/08/2026
The overreaction to this Hugging Face thing is driving me nuts. Come flay me for being wrong but... 1/x
14436126
Reposted by Fabio E. Tonti
Jacob T. Levy @jacobtlevy.bsky.social · 28/08/2026
Editors who at least *might* demand revisions if they saw a piece committing an obvious logical fallacy like affirming the consequent routinely *don't* ask pop psych or business psych contributors to demand revisions for selection on the dependent variable.
12219
Fabio E. Tonti @tontief.bsky.social · 11/08/2026
Sometimes have to repost to make everyone *worthy* feel less alone :D
010
Reposted by Fabio E. Tonti
Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 10/08/2026
Do believe we are near peak AI-slop in folks professional work. You see an LLM presentation and think not "wow, how pretty" but "oh, they must not have tried very hard"
5964
Reposted by Fabio E. Tonti
Thomas Lumley @tslumley.bsky.social · 05/08/2026
This doesn't work for confidence intervals that can be the whole real line. Ratio of Normal means (eg cost-effectiveness) is one genuine case (when both numerator and denominator have uncertain sign) and a trivial case is empty set with prob alpha and whole real line otherwise
131
Fabio E. Tonti @tontief.bsky.social · 02/08/2026
Exactly this!
010
Reposted by Fabio E. Tonti
Dorian Lynskey @dorianlynskey.bsky.social · 28/07/2026
I've written a few books and checked all the quotes (deleting some bangers because I doubted their authenticity) and if you're letting fabricated ones slip through then you're cutting corners in an egregious way. It is not the copy-editor's job to check every source, it's yours. Do the damn work
2640281
Reposted by Fabio E. Tonti
Ben Recht @beenwrekt.bsky.social · 27/07/2026
Now that the entire tech industry is on board, here’s a call to build AI models not just with open weights but with an open corpus.
argmin.net
Public Intelligence
A call for AI models with open weights, open source, and open corpus.
05311
Reposted by Fabio E. Tonti
Ben Recht @beenwrekt.bsky.social · 24/07/2026
Adding fuel to the fire, here's a post I wrote about open corpus models last July. It's all still true. www.argmin.net/p/an-open-mi...
argmin.net
An open mindset
The commitments required for fully open source machine learning
12910
Fabio E. Tonti @tontief.bsky.social · 24/07/2026
Not sure about this. There is a middle class (in many countries), studying in higher ed not simply to survive. Of course there are socio-econ pressures, but there are also other socio-cultural tendencies that push people into opportunism at all costs.
110
Fabio E. Tonti @tontief.bsky.social · 24/07/2026
I'd even say *rarely* instead of not always. Also, not only psych and bio, I feel like many engineering disciplines do that too, albeit with a slightly different focus. Classic in engineering: use "Analysis 1" for thinning stud population, more applied parts for "real stuff". Bad bad bad.
000
Reposted by Fabio E. Tonti
Gro-Tsen @gro-tsen.bsky.social · 20/07/2026
Let me close this thread by linking once more to the closest thing the mathematical community has to a collective position on AI and use of AI in mathematics, and the threats they pose: the Leiden Declaration leidendeclaration.ai •38/38
leidendeclaration.ai
Leiden Declaration on Artificial Intelligence and Mathematics
This declaration calls for action to address the challenges posed by the use of artificial intelligence within mathematics research.
4286
Fabio E. Tonti @tontief.bsky.social · 11/07/2026
That is very sad. I only knew him from some delightful email interactions and social media. And of course his books.
020
Reposted by Fabio E. Tonti
Lisa Messeri @lmesseri.bsky.social · 09/07/2026
It feels a certain way knowing that @mjcrockett.bsky.social and I published the rebuttal to this article… six months ago.
sciencedirect.com
AI Surrogates and illusions of generalizability in cognitive science
Recent advances in artificial intelligence (AI) have generated enthusiasm for using AI simulations of human research participants to generate new know…
27231
Reposted by Fabio E. Tonti
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 09/07/2026
The correlation coefficient is not a proper scoring rule. An agent who ranks models by correlation can be beaten by an infinite class of alternative rules. Here's a 2007 article on proper scoring rules: sites.stat.washington.edu/raftery/Rese...
310724
Reposted by Fabio E. Tonti
Olivia Guest · Ολίβια Γκεστ @olivia.science · 07/07/2026
I hate this feeling
5505
Reposted by Fabio E. Tonti
Per Engzell @pengzell.bsky.social · 05/07/2026
Friends don’t ask friends to read and comment on what is evidently Claude’s work
2493
Fabio E. Tonti @tontief.bsky.social · 04/07/2026
Cannot stop laughing.
110
Fabio E. Tonti @tontief.bsky.social · 02/07/2026
Looking forward to reading it. Whenever anyone posts something sensible about multiple testing on bsky, you always see the Bayesians coming out of the underscrub...
010
Fabio E. Tonti @tontief.bsky.social · 02/07/2026
A bit further in the article it basically says "doubly robust stuff + ipw GO BRRRRR" It makes me wonder whether Claude/gpt models would suggest that. Prolly. Do I want to know?
120
Reposted by Fabio E. Tonti
Darren Dahly @statsepi.bsky.social · 20/06/2026
I wonder if people dismiss these warnings because they implicitly assume that if these vehicles were really *that* dangerous, then obviously government would regulate them appropriately.
3112
Reposted by Fabio E. Tonti
Thomas Lumley @tslumley.bsky.social · 16/06/2026
I wrote a thing about the simple case of the a.s. representation theorem, which is basically about random number generation notstatschat.rbind.io/2026/06/16/t...
notstatschat.rbind.io
The simple case of almost sure representations - Biased and Inefficient
042
Reposted by Fabio E. Tonti
Carl T. Bergstrom @carlbergstrom.com · 03/06/2026
"The increasing involvement of technology companies in mathematical research raises the risk that research questions may come to be prioritized because of their amenability to automated mathematics, rather than expert judgment of their deeper significance."
leidendeclaration.ai
Leiden Declaration on Artificial Intelligence and Mathematics
This declaration calls for action to address the challenges posed by the use of artificial intelligence within mathematics research.
5306112
Reposted by Fabio E. Tonti
Darren Dahly @statsepi.bsky.social · 26/05/2026
Book: Test-Driven Data Analysis (by @njr0.bsky.social) Chapter 1 Data analysis as if the answers actually matter ht @cghlewis.bsky.social cc @rmcelreath.bsky.social book.tdda.info/book/chapter...
book.tdda.info
1 Orientation
13311
Reposted by Fabio E. Tonti
Nick Brown @steamtraen.eu · 09/05/2026
These are people who have never done anything like that in their lives. Their entire upbringing consisted of "nothing is ever your fault, it's always some nameless other".
0142
Reposted by Fabio E. Tonti
Christian Wolf @chriswolfvision.bsky.social · 05/05/2026
Who needs Netflix when all the drama is on Bluesky...
1234
Reposted by Fabio E. Tonti
Darren Dahly @statsepi.bsky.social · 02/05/2026
Looking forward to this @royalstatsoc.bsky.social discussion meeting on July 1 featuring this paper on randomisation by @stephensenn.bsky.social. Paper: rss.org.uk/RSS/media/Fi... Please register: rss.org.uk/training-eve...
rss.org.uk
082
Reposted by Fabio E. Tonti
sducrocq.bsky.social @sducrocq.bsky.social · 15/04/2026
​Happy 20th Anniversary to #rdatatable! 🎂 ​Since 2006, DT[i, j, by] has been the gold standard for high-performance data manipulation in R. Speed, efficiency, and elegance! ​A huge thanks to Matt Dowle, Arun Srinivasan, and all the contributors! @rdatatable.bsky.social 🥂 ​#rstats #rdatatable
0124
Fabio E. Tonti @tontief.bsky.social · 28/04/2026
Nice and well-motivated exposition! One thing: I'd replace Wolfram something with Sympy or similar.
040
Fabio E. Tonti @tontief.bsky.social · 18/04/2026
I was thinking exactly that when seeing those replies...
052
Reposted by Fabio E. Tonti
Ciaran Martin @ciaranm.bsky.social · 14/04/2026
They conducted a full test of its hacking prowess They evaluated its capabilities against current cyber defences And they published the whole evaluation in just over ONE week It is the only independent evaluation of this powerful new AI tool yet undertaken anywhere in the world 10/
2599
Reposted by Fabio E. Tonti
Georgia Tomova @georgiatomova.bsky.social · 13/04/2026
Recording now available if you've missed it:
youtu.be
Natalia Ortega | Meta-analyses of dietary exposures must consider energy adjustment
YouTube video by the Causal Inference Interest Group (CIIG)
0146
Reposted by Fabio E. Tonti
Jesspooka Kant 🎃 @jessdkant.bsky.social · 10/10/2025
Imagine for a moment a world so dependent on data centers that every aspect of municipal infrastructure must be reshaped. Not around human need, but maintaining GPU farms that use exponentially more power and water with each build. Models are only growing, and our resources are only diminishing.
351396
Reposted by Fabio E. Tonti
Darren Dahly @statsepi.bsky.social · 07/04/2026
Register 👇 events.teams.microsoft.com/event/8b7bd5...
Wednesday, June 3, 2:00 PM
Upcoming
NIHR RSS Research Methods Academy: A Practical Introduction to Bayesian Methods for Clinical Research
074
Fabio E. Tonti @tontief.bsky.social · 06/04/2026
I wish some people on here gave recommendations more often:)
020