Esther Mondragón
@emp1.bsky.social
Hic sunt dracones
I represent myself, no one else.
Computational and associative learning.
Biologically-inspired AI.
She/They
https://cit-ai.net/people.html#Mondragon
https://www.city.ac.uk/about/people/academics/esther-mondragon
I represent myself, no one else.
Computational and associative learning.
Biologically-inspired AI.
She/They
https://cit-ai.net/people.html#Mondragon
https://www.city.ac.uk/about/people/academics/esther-mondragon
Pinned
Esther Mondragón
@emp1.bsky.social
· Jun 9
Advancing the Biological Plausibility and Efficacy of Hebbian Convolutional Neural Networks
The research presented in this paper advances the integration of Hebbian learning into Convolutional Neural Networks (CNNs) for image processing, syst…
www.sciencedirect.com
Ablation study and resulting optimal architecture that considerably improves recent research on CNN-Hebbian learning integration with competition mechanisms.
with @julian-jn.bsky.social
www.sciencedirect.com/science/arti...
with @julian-jn.bsky.social
www.sciencedirect.com/science/arti...
Behaviourism, regarded as an aversive US, is now paired with other still more neutral terms in an attempt to transfer its negative value to those labels. I wonder whether Pavlov would have approved; Watson would most certainly have done so. 🤔
You cannot expect me not to point this out, right?
You cannot expect me not to point this out, right?
a brown and white dog is sitting next to a man on a couch and looking at the camera .
ALT: a brown and white dog is sitting next to a man on a couch and looking at the camera .
media.tenor.com
February 15, 2026 at 3:19 PM
Behaviourism, regarded as an aversive US, is now paired with other still more neutral terms in an attempt to transfer its negative value to those labels. I wonder whether Pavlov would have approved; Watson would most certainly have done so. 🤔
You cannot expect me not to point this out, right?
You cannot expect me not to point this out, right?
Reposted by Esther Mondragón
This piece is about CS conferences, but it speaks to a larger trend. What are we, as individual researchers and scholarly communities, doing to prevent epistemic poison? What can we do to preserve and maintain the quality of our knowledge in this era of bullshit and slop?
⚗️ 🧪 #SciComm #Ethics
⚗️ 🧪 #SciComm #Ethics
A review of the proceedings from four major computer-science conferences showed that none from 2021, and all from 2025, had fake citations.
arxiv.org/abs/2602.058...
#AI #LLMs #Hallucinations #Misconduct #ScholComm
arxiv.org/abs/2602.058...
#AI #LLMs #Hallucinations #Misconduct #ScholComm
The Case of the Mysterious Citations
Mysterious citations are routinely appearing in peer-reviewed publications throughout the scientific community. In this paper, we developed an automated pipeline and examine the proceedings of four ma...
arxiv.org
February 15, 2026 at 11:49 AM
Enough is enough, Mr Taranis! ☔
www.metoffice.gov.uk/blog/2026/me...
www.metoffice.gov.uk/blog/2026/me...
Deep Dive: Understanding this winter’s remarkably persistent weather pattern
This winter has brought an extraordinary spell of weather across the UK, marked by striking contrasts, record‑breaking rainfall in some regions and an unusually stubborn atmospheric pattern that has r...
www.metoffice.gov.uk
February 15, 2026 at 10:21 AM
Enough is enough, Mr Taranis! ☔
www.metoffice.gov.uk/blog/2026/me...
www.metoffice.gov.uk/blog/2026/me...
Reposted by Esther Mondragón
I found one of the chapters of Mary Henle’s book online, previously published as article, this one:
Henle, M. (1978). One man against the Nazis: Wolfgang Köhler. American Psychologist, 33(10), 939.
Let me see if I can find some of the other chapters too.
pure.mpg.de/rest/items/i...
6/🧵
Henle, M. (1978). One man against the Nazis: Wolfgang Köhler. American Psychologist, 33(10), 939.
Let me see if I can find some of the other chapters too.
pure.mpg.de/rest/items/i...
6/🧵
February 14, 2026 at 6:50 PM
I found one of the chapters of Mary Henle’s book online, previously published as article, this one:
Henle, M. (1978). One man against the Nazis: Wolfgang Köhler. American Psychologist, 33(10), 939.
Let me see if I can find some of the other chapters too.
pure.mpg.de/rest/items/i...
6/🧵
Henle, M. (1978). One man against the Nazis: Wolfgang Köhler. American Psychologist, 33(10), 939.
Let me see if I can find some of the other chapters too.
pure.mpg.de/rest/items/i...
6/🧵
Reposted by Esther Mondragón
even tho these examples are so tired, the reliance on these systems grows... it's so messed up
It’s so funny to think about the fact that there’s people out there deferring every life decision to this
February 14, 2026 at 1:19 PM
even tho these examples are so tired, the reliance on these systems grows... it's so messed up
I very much miss watching the waves here (not sure about this time though) Luckily, there is always a friend who sends a video. 🌊
February 14, 2026 at 3:50 PM
I very much miss watching the waves here (not sure about this time though) Luckily, there is always a friend who sends a video. 🌊
Reposted by Esther Mondragón
Bc it means certain people cannot play only by their own rules to determine who's in their club, plays their game, or who isn't. Feminization means democratization. It means access & sharing with others. It is skill-based participation, not hand-shake agreement. And some people cannot have that.
Putting aside all other nonsense spat out by these terrified fools, what is the problem with the "feminisation of higher education"? Women make up over 50% of the UK population and nearly 50% of the world's population. Why should male views be more preferred over those of women? ✂️🪄
February 14, 2026 at 2:38 PM
Bc it means certain people cannot play only by their own rules to determine who's in their club, plays their game, or who isn't. Feminization means democratization. It means access & sharing with others. It is skill-based participation, not hand-shake agreement. And some people cannot have that.
Reposted by Esther Mondragón
It's also not unrelated to this logic and behaviour bsky.app/profile/sexa...
Watching pro AI people in Reddit discussions about slop PRs (code edit submissions to open source public software projects) complaining "What's the problem just review the PRs, if they're good merge them" really underlines a lot of the problematic way that AI boosters devalue respect for labour.
February 14, 2026 at 2:17 PM
It's also not unrelated to this logic and behaviour bsky.app/profile/sexa...
Putting aside all other nonsense spat out by these terrified fools, what is the problem with the "feminisation of higher education"? Women make up over 50% of the UK population and nearly 50% of the world's population. Why should male views be more preferred over those of women? ✂️🪄
February 14, 2026 at 10:52 AM
Putting aside all other nonsense spat out by these terrified fools, what is the problem with the "feminisation of higher education"? Women make up over 50% of the UK population and nearly 50% of the world's population. Why should male views be more preferred over those of women? ✂️🪄
February 14, 2026 at 10:30 AM
Reposted by Esther Mondragón
Whether you agree with all of the below or not, Berna is imo the person doing the most interesting thinking in the metascience space. It’s provocative (cuts against the accepted wisdom) but very important.
I do really strongly agree with her about the crisis/disaster narrative.
I do really strongly agree with her about the crisis/disaster narrative.
On the publication bias discourse, I regret that metascience has become a source of decontextualized, low-res, bean-counting-focused `science is in crisis' narratives. It is largely uncurious abt science, desperately lacking in theory & measurement. I'll quote a few takes I liked & add my thoughts🧵
February 13, 2026 at 8:43 AM
Whether you agree with all of the below or not, Berna is imo the person doing the most interesting thinking in the metascience space. It’s provocative (cuts against the accepted wisdom) but very important.
I do really strongly agree with her about the crisis/disaster narrative.
I do really strongly agree with her about the crisis/disaster narrative.
Reposted by Esther Mondragón
On the publication bias discourse, I regret that metascience has become a source of decontextualized, low-res, bean-counting-focused `science is in crisis' narratives. It is largely uncurious abt science, desperately lacking in theory & measurement. I'll quote a few takes I liked & add my thoughts🧵
February 13, 2026 at 12:41 AM
On the publication bias discourse, I regret that metascience has become a source of decontextualized, low-res, bean-counting-focused `science is in crisis' narratives. It is largely uncurious abt science, desperately lacking in theory & measurement. I'll quote a few takes I liked & add my thoughts🧵
Reposted by Esther Mondragón
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
PALMS: Pavlovian Associative Learning Models Simulator
Simulations are an indispensable step in the cycle of theory development and refinement, helping researchers formulate precise definitions, generate models, and make accurate predictions. This paper i...
arxiv.org
February 10, 2026 at 2:49 PM
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
Modern times: Got a bit of good news, and now I am worrying when and where the slap will come from.
a close up of a monkey looking at the camera
ALT: a close up of a monkey looking at the camera
media.tenor.com
February 11, 2026 at 7:31 PM
Modern times: Got a bit of good news, and now I am worrying when and where the slap will come from.
Reposted by Esther Mondragón
We are demanding that EU co-legislators reject attempts in the AI Omnibus to remove a key transparency safeguard from the AI Act.
We cannot open a loophole that would let providers exempt themselves from the AI Act’s high risk requirements with no transparency www.accessnow.org/press-releas...
We cannot open a loophole that would let providers exempt themselves from the AI Act’s high risk requirements with no transparency www.accessnow.org/press-releas...
Access Now - A call to EU legislators: protect rights and reject the call to delete transparency safeguard in AI Act
We, the undersigned organisations and individuals, urge you in the strongest possible terms to reject the deletion of the Article 49(2) transparency safeguard for high-risk AI systems that is proposed...
www.accessnow.org
February 11, 2026 at 9:47 AM
We are demanding that EU co-legislators reject attempts in the AI Omnibus to remove a key transparency safeguard from the AI Act.
We cannot open a loophole that would let providers exempt themselves from the AI Act’s high risk requirements with no transparency www.accessnow.org/press-releas...
We cannot open a loophole that would let providers exempt themselves from the AI Act’s high risk requirements with no transparency www.accessnow.org/press-releas...
Reposted by Esther Mondragón
The relevant video about how LLMs impact human learning is below in case anyone is curious and hasn't seen it already (it's part of a 7-part series which is my pinned tweet and also findable from YouTube if you go to my playlists)
Part 4: How do LLMs impact learning?
YouTube video by Andrew Perfors
youtu.be
February 10, 2026 at 9:18 PM
The relevant video about how LLMs impact human learning is below in case anyone is curious and hasn't seen it already (it's part of a 7-part series which is my pinned tweet and also findable from YouTube if you go to my playlists)
Reposted by Esther Mondragón
A very nice resource to have - Particularly happy to see it confirmed that my recent experiments on latent inhibition (and its reversal) cause a significant headache for Pearce-Hall and Hybrid models.
psycnet.apa.org/record/2025-...
psycnet.apa.org/record/2025-...
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
PALMS: Pavlovian Associative Learning Models Simulator
Simulations are an indispensable step in the cycle of theory development and refinement, helping researchers formulate precise definitions, generate models, and make accurate predictions. This paper i...
arxiv.org
February 10, 2026 at 5:22 PM
A very nice resource to have - Particularly happy to see it confirmed that my recent experiments on latent inhibition (and its reversal) cause a significant headache for Pearce-Hall and Hybrid models.
psycnet.apa.org/record/2025-...
psycnet.apa.org/record/2025-...
Reposted by Esther Mondragón
rock paper scissors
'AI, Ethics and Society' sounds like a dataset for contrastive learning 🤣
February 10, 2026 at 3:51 PM
rock paper scissors
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
PALMS: Pavlovian Associative Learning Models Simulator
Simulations are an indispensable step in the cycle of theory development and refinement, helping researchers formulate precise definitions, generate models, and make accurate predictions. This paper i...
arxiv.org
February 10, 2026 at 2:49 PM
New preprint and simulator of associative learning attentional models. Have fun! 👁️
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
arxiv.org/abs/2602.07519
cal-r.org/index.php?id...
#simulation #associative_learning #attention
Reposted by Esther Mondragón
February 9, 2026 at 2:34 PM
Reposted by Esther Mondragón
"In summary, our results suggest that theoretically interesting connections between LLMs and brains on three neural datasets are driven largely by fragile methodologies and overlooked confounds." -- who would have thonked that?
neuroAI comparisons of ANNs to brains do have a range of problems. Even more than I had realized. And I was worried before: www.biorxiv.org/content/10.1...
www.biorxiv.org
February 9, 2026 at 4:39 PM
"In summary, our results suggest that theoretically interesting connections between LLMs and brains on three neural datasets are driven largely by fragile methodologies and overlooked confounds." -- who would have thonked that?
February 9, 2026 at 2:34 PM
Reposted by Esther Mondragón
' Ronald Barnett describes the contemporary fascination with skills as a form of ‘bewitchment’ (others have put it more strongly elsewhere)....When individuals are defined in this way, they become more easily interchangeable in the labour market, increasing workers’ vulnerability.'
How did skills take over higher education? One short history - HEPI
Over the weekend HEPI published blogs on the freezing of student loan thresholds, and the Westminster Hall debates on duty of care. This blog was kindly authored by Dr. Josh Patel, Senior Education an...
www.hepi.ac.uk
February 9, 2026 at 8:54 AM
' Ronald Barnett describes the contemporary fascination with skills as a form of ‘bewitchment’ (others have put it more strongly elsewhere)....When individuals are defined in this way, they become more easily interchangeable in the labour market, increasing workers’ vulnerability.'
Reposted by Esther Mondragón
The big consciousness conference in Tucson/Arizona is cancelled. Philosophy & Cognitive Science are fields long known to be largely inside-handshake dudebro clubs, something which I said out loud (it's not that they'd hired me), and now revealed to be disproportionally present in the Epstein files.
February 7, 2026 at 2:43 PM
The big consciousness conference in Tucson/Arizona is cancelled. Philosophy & Cognitive Science are fields long known to be largely inside-handshake dudebro clubs, something which I said out loud (it's not that they'd hired me), and now revealed to be disproportionally present in the Epstein files.
Reposted by Esther Mondragón
No joke: I got angry hate mail today for writing an obituary of a Black woman scientist—because the person felt she did didn’t deserve the recognition.
Which just makes me want to share it again: www.nature.com/articles/d41...
Which just makes me want to share it again: www.nature.com/articles/d41...
Gladys Mae West obituary: mathematician who pioneered GPS technology
She made key contributions to US cold-war science despite facing huge barriers as a Black woman.
www.nature.com
February 6, 2026 at 9:09 AM
No joke: I got angry hate mail today for writing an obituary of a Black woman scientist—because the person felt she did didn’t deserve the recognition.
Which just makes me want to share it again: www.nature.com/articles/d41...
Which just makes me want to share it again: www.nature.com/articles/d41...