Sign in

Rui Chaves

@rpchaves.bsky.social
98 followers 36 following 60 posts

"Words, words, words." - Hamlet, Act II, scene ii. (he/him)

PostsRepliesMedia
Rui Chaves @rpchaves.bsky.social · 07/10/2026
Will the US ever become a direct democracy like all others ? The Electoral College fails its purpose. In 1968 an amendment to abolish EC was approved by the House but failed to move past the Senate. More resolutions have been submitted to amend the ElC than any other part of the constitution
000
Rui Chaves @rpchaves.bsky.social · 29/09/2026
FWIW, a more literal translation: “only he who yearns for the absurd is capable of conquering the impossible “.
042
Rui Chaves @rpchaves.bsky.social · 08/09/2026
This was a matter of time, but it is not clear which brands will dominate and conquer more market share. Will Canada embrace these brands? Interesting to see.
001
Rui Chaves @rpchaves.bsky.social · 08/09/2026
Chinese cars currently sold in Europe, including EVs/HVs, are substantially cheaper because labor costs less and many parts are created in China already. They pass strict EU safety standards and are selling well. But there are so many brands that technical support and part replacement is tricky.
100
Rui Chaves @rpchaves.bsky.social · 06/09/2026
<sarcasm> It's not a war according to the government: apnews.com/article/iran... Imagine Canada dropping a couple of bombs on a US city. Nobody would take it as a declaration of war either, right? Just a little conflict. </sarcasm>
apnews.com
Vance says Iran fight isn't a 'war' as Trump tries to navigate unpopular conflict as election nears
Vice President JD Vance says he wouldn't call the U.S. conflict with Iran a “war,” despite recent escalations.
010
Rui Chaves @rpchaves.bsky.social · 27/08/2026
aiandeducation.mit.edu/report/#26-n...
aiandeducation.mit.edu
Report – AI and Education
000
Rui Chaves @rpchaves.bsky.social · 30/07/2026
He is catering to his audience, defunding 'useless' and 'woke' research. It's just a distraction. But one that will set back the US for decades, and cost lives.
020
Rui Chaves @rpchaves.bsky.social · 22/07/2026
"AI" has been verbed: "How many people are going to A.I. the heck out of this photo (...)" [x.com/AldenOnTheAir/status/20780987…] "We are not going to AI the job of a nurse" [www.youtube.com/watch?v=XJ508FyLQq4] "You should AI that with copilot." (Copilot advertisement)
000
Rui Chaves @rpchaves.bsky.social · 17/07/2026
Anyway, the dip in the raw job posts is hinted by the shared area plots. Plus, the GitHub repo has always contained the script to extract the raw job counts.
100
Rui Chaves @rpchaves.bsky.social · 17/07/2026
It’s not clear these data can be used for forecasting. Things are bound to be very different in 5 years, judging by data going back to the 90s, so there is a sense that current data trends can be misleading in the long run.
100
Rui Chaves @rpchaves.bsky.social · 14/07/2026
Very nice! I opted to not openly report the raw job count plot because it is very (too?) depressing. Many posts are for Applied Linguistics, actually, which is often about Language Teaching rather than linguistics per se.
110
Rui Chaves @rpchaves.bsky.social · 10/06/2026
The US could probably make higher education free. Wouldn't that be in the interest of national security?
000
Reposted by Rui Chaves
Our World in Data @ourworldindata.org · 10/06/2026
How much do countries spend on their militaries? How has this changed over time, and how do different countries compare? The Stockholm International Peace Research Institute (@sipri.org) provides a key resource for answering these questions and more.
Small multiples line charts of military spending by country where they compare annual spending from 1949 to about 2026 in constant 2024 US dollars to show long-term trends. Source: Stockholm International Peace Research Institute (2026). License: CC BY to Our World in Data.
32911
Rui Chaves @rpchaves.bsky.social · 03/06/2026
www.yahoo.com/news/world/a...
yahoo.com
China tries to balances AI push with job displacement fears
China is trying to balance its society-wide AI push with mounting concerns over the tech’s disruptive socioeconomic effects.
000
Rui Chaves @rpchaves.bsky.social · 18/05/2026
Europe and China can continue this work?
000
Rui Chaves @rpchaves.bsky.social · 01/05/2026
This! www.caixinglobal.com/2026-04-30/c...
caixinglobal.com
Chinese Courts Rule Companies Cannot Fire Workers Simply to Replace Them With AI
Judges classify AI adoption as a controllable business strategy rather than an unavoidable disruption, shielding employees from automation-driven layoffs
110
Rui Chaves @rpchaves.bsky.social · 15/04/2026
b) why on Earth would the order of the disjuncts be expected to be plural < singular?? Complete nonsense. If only there was a field of study that could inform school teachers about basic facts of English... if there was, they could take such a class in college, and better serve kids. Oh well...
030
Rui Chaves @rpchaves.bsky.social · 15/04/2026
My 10yo is going over a 5th grade study guide. The system tells him that "Either the students or the teacher is late" is correct, but "Either the teacher or the students are late" is incorrect. This is a double whammy of wrong: (a) the Proximity Rule (verb agrees with last disjunct) is made-up rule.
120
Rui Chaves @rpchaves.bsky.social · 11/04/2026
Now the really hard part: get them out of the capsule.
000
Reposted by Rui Chaves
🇵🇸Tim Henke (tɪm 'ɦɛŋ.kə) @timhenke.bsky.social · 28/03/2026
Very amusing vowel space playground editor.p5js.org/mimoi/full/x...
67925
Rui Chaves @rpchaves.bsky.social · 28/03/2026
P5 is my favorite! So cool!
010
Rui Chaves @rpchaves.bsky.social · 15/01/2026
Can we improve the road signs near border accesses? People make a wrong turn too often, and now there's a steeper price to pay: www.wgrz.com/article/news...
wgrz.com
ICE detains UB researcher after his attempted return from Canada
A University at Buffalo medical researcher was taken into custody after attempting to re-enter the United States earlier this month.
000
Rui Chaves @rpchaves.bsky.social · 01/01/2026
All code and more plots and regressions at: github.com/RuiPChaves/L...
010
Rui Chaves @rpchaves.bsky.social · 01/01/2026
Number of Linguistics job postings have dropped significantly in the Linguist List. Please note that the numbers reported below are inflated because some job postings mention multiple areas of study (e.g. phonology + phonetics, or computational linguistics + semantics).
141
Rui Chaves @rpchaves.bsky.social · 30/10/2025
Putting my Asimov hat on, maybe one day tutor systems will be very good, at least for some people. But I can’t imagine being a kid and enjoying learning from a screen rather than from a person, in a spontaneous classroom environment. Most of us are social creatures.
130
Rui Chaves @rpchaves.bsky.social · 30/10/2025
The deeper the model goes into a topic the worse it gets (plain wrong or hallucinates refs), and I suspect it depends a lot on the field. Including when it is asked to create exercises. For basic things probably fine, but how can a student tell when the model strays from the truth?
1150
Rui Chaves @rpchaves.bsky.social · 14/10/2025
420k malicious tokens (0.00016% of the total training tokens) were enough to ruin the model, even after fine-tuning. This means the public can attack these models by creating and uploading bad data to Github, Stackoverflow, blogs, etc...
010
Rui Chaves @rpchaves.bsky.social · 14/10/2025
Yet another example of how LLMs are not robust (a key argument against symbolic AI): a small fraction of 'bad' data can compromise models regardless of dataset or model size. Data poisoning does not scale with model size. arxiv.org/pdf/2510.07192
arxiv.org
110
Rui Chaves @rpchaves.bsky.social · 20/07/2025
www.youtube.com/watch?v=tCsl...
youtube.com
The Physics Of Dissonance
YouTube video by minutephysics
010
Rui Chaves @rpchaves.bsky.social · 09/07/2025
At Starbucks I order a Short coffee. I suspect most people don’t even know that exists…
010
Rui Chaves @rpchaves.bsky.social · 30/06/2025
There is mounting evidence that LLMs rely heavily on brute-force and shallow heuristics to achieve high accuracy, despite some studies (though not all) finding evidence for sophisticated latent representations. Add another to the bunch: www.arxiv.org/abs/2506.16678
arxiv.org
Mechanisms vs. Outcomes: Probing for Syntax Fails to Explain Performance on Targeted Syntactic Evaluations
Large Language Models (LLMs) exhibit a robust mastery of syntax when processing and generating text. While this suggests internalized understanding of hierarchical syntax and dependency relations, the...
021
Rui Chaves @rpchaves.bsky.social · 24/06/2025
Coolest? It's a tie between Braitenberg's "Vehicles" and Hofstadter "GEB".
mitpress.mit.edu
Vehicles
These imaginative thought experiments are the inventions of one of the world's eminent brain researchers. They are "vehicles," a series of hypothetical, self...
000
Rui Chaves @rpchaves.bsky.social · 16/06/2025
He has never read a book in his life.
020
Rui Chaves @rpchaves.bsky.social · 12/06/2025
I feel sick.
030
Rui Chaves @rpchaves.bsky.social · 10/06/2025
We even got real close to curing HIV: www.iavi.org/press-releas.... But this uses mrna, and that has come to be a dirty word for these people. It shouldn't. It's a triumph of modern molecular biology.
iavi.org
Two HIV vaccine trials show proof of concept for pathway to broadly neutralizing antibodies
Learn more about North American and African clinical trials led by Scripps Research, IAVI and collaborators that show early immune activation using mRNA vaccine technology—marking progress toward an e...
110
Reposted by Rui Chaves
Ada Palmer @adapalmer.bsky.social · 03/06/2025
Wow, yes. An amazing glimpse of the LLM Lie Machine at work. Anyone whose students are bring tempted by LLMs, sharing this could be very helpful showing how it lies, and how it fails.
12269137
Rui Chaves @rpchaves.bsky.social · 02/06/2025
I'm so sorry. Beyond words.
010
Reposted by Rui Chaves
Kimberly Quinn 🇨🇦 @kimberlyquinn.bsky.social · 17/05/2025
"So here’s one place my compass is pointing me to: pushing back against bullshit. Frequently. Publicly. This frightens me because I have enough to deal with...I’m a people pleaser. I love being loved. But we are in terrifying times and all I have is my compass." - @jimmycoan.bsky.social
jimcoan.substack.com
Science Is Fraught
Ignorance can't govern and reality isn't negotiable
1499
Rui Chaves @rpchaves.bsky.social · 08/05/2025
Bresnan (1991) journals.linguisticsociety.org/proceedings/...
022
Reposted by Rui Chaves
Laura A. Michaelis @congramqueen.bsky.social · 11/03/2025
Register now for the LSA Linguistic Institute in Eugene, OR: over 80 courses over two 2.5-week terms (July 7-22::July 24-August 8). I'm teaching Constructions & the Grammar of Context with E Francis in session 1! center.uoregon.edu/LSA/2025/
center.uoregon.edu
Linguistics Society of America Summer Institute
Linguistics Society of America Summer Institute
032
Rui Chaves @rpchaves.bsky.social · 05/03/2025
Interesting! I've no problem with that. This is crafted as a cartoon, so it is read top to bottom.
010
Rui Chaves @rpchaves.bsky.social · 06/02/2025
Thank you for doing this work, it is very interesting! Challenge: I suspect if you asked o1 to create an X-bar grammar sufficiently large to parse a couple of pages of English it would not do a good job .
010
Rui Chaves @rpchaves.bsky.social · 06/02/2025
I've annotated the o1 trees here: drive.google.com/file/d/1we0S... I think these systems *are* amazing, and getting better and better, but have a long way to go. Major issues: inconsistent analyses (suggesting not a single grammar exists), wrong analysis, partial trees, and gibberish trees (only 2).
drive.google.com
o1.pdf
200
Rui Chaves @rpchaves.bsky.social · 06/02/2025
O1's results are impressive, but they are not as good as a professional human annotator, at least for these sentences.
110
Rui Chaves @rpchaves.bsky.social · 06/02/2025
This is a good one but the other outputs form o1, are not good, and often mutually inconsistent. E.g. in the "This town's only road's potholes' depths are driving me insane" example the VP is too flat, the "only road" NP is wrong, inconsistent with tree for "a medium-sized sturdy lamp" in next S
110
Rui Chaves @rpchaves.bsky.social · 06/02/2025
Maybe they are assuming AI will bring about those breakthroughs? What happens when that bubble eventually pops?
010
Rui Chaves @rpchaves.bsky.social · 07/01/2025
ps - Should have used NDL instead of HOL.
000
Rui Chaves @rpchaves.bsky.social · 06/01/2025
The LL mostly consists of US posts. Here are the 2024 posts for "Professor":
120
Rui Chaves @rpchaves.bsky.social · 05/01/2025
That's a great idea. Maybe mine ProQuest, LingBuzz, and similar venues?
120
Rui Chaves @rpchaves.bsky.social · 05/01/2025
Keep in mind that there are hardly any CL posts that don't mention another discipline, so that blue bar mostly consists of posts from a few of the other bars. Similarly, some posts mention both phonetics and phonology, or syntax and semantics, etc. The raw post count is even smaller.
120