Prof. Evelina Leivada
LightNews LightNews
banner
evelinaleivada.bsky.social
Prof. Evelina Leivada
@evelinaleivada.bsky.social
210 followers 79 following 9 posts
ICREA Research Professor | Psycholinguist @icreacommunity & @UABBarcelona 🧠🗣 | Open Access advocate ✍ | Assoc. Editor @biolinguistics 🤹‍♀️ | ☆ Alba's & Laura's mom ☆
Posts Replies Media Videos
Reposted by Prof. Evelina Leivada
olivia.science
Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...

We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
Abstract: Under the banner of progress, products have been uncritically adopted or even imposed on users — in past centuries with tobacco and combustion engines, and in the 21st with social media. For these collective blunders, we now regret our involvement or apathy as scientists, and society struggles to put the genie back in the bottle. Currently, we are similarly entangled with artificial intelligence (AI) technology. For example, software updates are rolled out seamlessly and non-consensually, Microsoft Office is bundled with chatbots, and we, our students, and our employers have had no say, as it is not considered a valid position to reject AI technologies in our teaching and research. This is why in June 2025, we co-authored an Open Letter calling on our employers to reverse and rethink their stance on uncritically adopting AI technologies. In this position piece, we expound on why universities must take their role seriously toa) counter the technology industry’s marketing, hype, and harm; and to b) safeguard higher education, critical thinking, expertise, academic freedom, and scientific integrity. We include pointers to relevant work to further inform our colleagues. Figure 1. A cartoon set theoretic view on various terms (see Table 1) used when discussing the superset AI (black outline, hatched background): LLMs are in orange; ANNs are in magenta; generative models are in blue; and finally, chatbots are in green. Where these intersect, the colours reflect that, e.g. generative adversarial network (GAN) and Boltzmann machine (BM) models are in the purple subset because they are both generative and ANNs. In the case of proprietary closed source models, e.g. OpenAI’s ChatGPT and Apple’s Siri, we cannot verify their implementation and so academics can only make educated guesses (cf. Dingemanse 2025). Undefined terms used above: BERT (Devlin et al. 2019); AlexNet (Krizhevsky et al. 2017); A.L.I.C.E. (Wallace 2009); ELIZA (Weizenbaum 1966); Jabberwacky (Twist 2003); linear discriminant analysis (LDA); quadratic discriminant analysis (QDA). Table 1. Below some of the typical terminological disarray is untangled. Importantly, none of these terms are orthogonal nor do they exclusively pick out the types of products we may wish to critique or proscribe. Protecting the Ecosystem of Human Knowledge: Five Principles
September 6, 2025 at 8:13 AM
Reposted by Prof. Evelina Leivada
garymarcus.bsky.social
GenAI hasn’t created memorable songs, books, & movies, despite vast amounts of input. Why?

1. Everything GA AI does is derivative & tends to be a kind of average.

2. It lacks a conceptual understanding of human experience.

3. Because GenAI lacks robust world models, its output lacks coherence.
May 4, 2025 at 10:55 PM
Reposted by Prof. Evelina Leivada
olivia.science
Tired but happy to say this is out w @andreaeyleen.bsky.social: Are Neurocognitive Representations 'Small Cakes'? philsci-archive.pitt.edu/24834/

We analyse cog neuro theories showing how vicious regress, e.g. the homunculus fallacy, is (sadly) alive and well — and importantly how to avoid it. 1/
In order to understand cognition, we often recruit analogies as building blocks of theories to aid us in this quest. One such attempt, originating in folklore and alchemy, is the homunculus: a miniature human who resides in the skull and performs cognition. Perhaps surprisingly, this appears indistinguishable from the implicit proposal of many neurocognitive theories, including that of the 'cognitive map,' which proposes a representational substrate for episodic memories and navigational capacities. In such 'small cakes' cases, neurocognitive representations are assumed to be meaningful and about the world, though it is wholly unclear who is reading them, how they are interpreted, and how they come to mean what they do. We analyze the 'small cakes' problem in neurocognitive theories (including, but not limited to, the cognitive map) and find that such an approach a) causes infinite regress in the explanatory chain, requiring a human-in-the-loop to resolve, and b) results in a computationally inert account of representation, providing neither a function nor a mechanism. We caution against a 'small cakes' theoretical practice across computational cognitive modelling, neuroscience, and artificial intelligence, wherein the scientist inserts their (or other humans') cognition into models because otherwise the models neither perform as advertised, nor mean what they are purported to, without said 'cake insertion.' We argue that the solution is to tease apart explanandum and explanans for a given scientific investigation, with an eye towards avoiding van Rooij's (formal) or Ryle's (informal) infinite regresses. Figure 1 in https://philsci-archive.pitt.edu/24834/ Box 1 in https://philsci-archive.pitt.edu/24834/ Box 2 in https://philsci-archive.pitt.edu/24834/
March 1, 2025 at 2:16 PM
evelinaleivada.bsky.social
New postdoc position open in my project TURING! If you are interested in Large Language Models, come work with us. Deadline 7 March!
euraxess.ec.europa.eu/jobs/318547
1 Post-Doctoral Researcher position (2025DILIFRUA20)
Job position  1 Post-Doctoral Researcher position Job position requirements:
euraxess.ec.europa.eu
February 19, 2025 at 3:07 PM
Reposted by Prof. Evelina Leivada
elliot-murphy.bsky.social
New paper out! 🚨🤖

We provide a comprehensive assessment of the newest reasoning model from OpenAI (o3-mini-high). We show that it fails to even remotely exhibit human-like linguistic (syntactic/compositional) competence. @garymarcus.bsky.social @evelinaleivada.bsky.social

arxiv.org/abs/2502.10934
Fundamental Principles of Linguistic Structure are Not Represented by o3
A core component of a successful artificial general intelligence would be the rapid creation and manipulation of grounded compositional abstractions and the demonstration of expertise in the family of...
arxiv.org
February 18, 2025 at 5:34 AM
Reposted by Prof. Evelina Leivada
irisvanrooij.bsky.social
🎬🎥🍿 Video of my keynote at MathPsych2024 now available online www.youtube.com/watch?v=WrwN...

#CogSci #CriticalAI #AIhype #AGI #PsychSci #PhilSci 🧪
Iris van Rooij keynote at MathPsych/ICCM 2024
YouTube video by Society for Mathematical Psychology
www.youtube.com
January 11, 2025 at 5:34 PM
Reposted by Prof. Evelina Leivada
tomerullman.bsky.social
thinking of calling this "The Illusion Illusion"

(more examples below)
December 1, 2024 at 2:33 PM
Reposted by Prof. Evelina Leivada
gretchenmcc.bsky.social
Hello! I'm an internet linguist!

I wrote a book called Because Internet about how we use language online gretchenmcculloch.com/book

I make @lingthusiasm.bsky.social, a podcast that's enthusiastic about linguistics

And I maintain a linguistics starter pack here: go.bsky.app/UUM7Gcx
November 8, 2024 at 5:06 PM
evelinaleivada.bsky.social
New paper out at Scientific Reports! www.nature.com/articles/s41...
Testing AI on language comprehension tasks reveals insensitivity to underlying meaning - Scientific Reports
Scientific Reports - Testing AI on language comprehension tasks reveals insensitivity to underlying meaning
www.nature.com
November 15, 2024 at 8:30 PM
evelinaleivada.bsky.social
One day I hope someone will develop the habit of reading the paper past p. 3. Because if one does it, one will know that we do test the ability to understand questions within a hmm given discourse context.
spiantado.bsky.social steven t. piantadosi @spiantado.bsky.social · Nov 15
One day I hope someone will explain why showing LLMs are bad on sentences that *people* famously do poorly on tests anything... and why "dogs dogs dog dog dogs" has much weight about their understanding of syntax relative to, you know, the ability to fluently write sentences and discourses
November 15, 2024 at 8:27 PM
Reposted by Prof. Evelina Leivada
haspelmath.bsky.social
25 workshops at SLE 2024 in Helsinki (deadline: 15 January): societaslinguistica.eu/sle2024/list...
List of workshops - 57th Annual Meeting of the Societas Linguistica Europaea
Calls on the Linguistlist WS1 Anglicism research in Europe: from vocabulary to use (Gisle Andersen, Elizabeth Peterson & Eline Zenner) WS2 Construction Grammar meets Sociolinguistics (Lotte Sommerer &...
societaslinguistica.eu
December 20, 2023 at 3:40 PM
evelinaleivada.bsky.social
🎈Great news!The Agencia Estatal de Investigación funded my project TURING (The Language Understanding of Artificial Intelligence Applications) with ca. 340K!! Elated for this new adventure! Stay tuned, I'll be soon hiring 2 or 3 post-docs. 🥳
December 19, 2023 at 9:55 AM
evelinaleivada.bsky.social
New paper in PNAS! Systematic testing of 3 Large Language Models reveals low language accuracy, absence of response stability, and a yes-response bias. We also tested humans in the exact same tasks and their performance as both stable and largely accurate. www.pnas.org/doi/10.1073/...
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
December 14, 2023 at 9:58 AM
Reposted by Prof. Evelina Leivada
stefanfrank.bsky.social
The Bilingual Dual-Path model: simulating bilingual production, comprehension, and development osf.io/preprints/ps.... A new preprint by @yunghankhoe.bsky.social, who's new to BlueSky and would like more followers
December 12, 2023 at 6:03 PM