Sign in

Pierre-Yves Oudeyer

@pyoudeyer.bsky.social
509 followers 456 following 71 posts

Research director @Inria, Head of @flowersInria lab, prev. @MSFTResearch @SonyCSLParis Artificial intelligence, cognitive sciences, sciences of curiosity, language, self-organization, autotelic agents, education, AI and society www.pyoudeyer.com

PostsRepliesMedia
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/09/2026
I'm happy to share a new podcast where I explain how I got to study curiosity, why it's fundamental in child development and how this got applied in self-improving robots/AI! 🙏 Thanks to the amazing Lynn Borton for hosting the podcast ! lynnborton.com/2026/08/27/m...
lynnborton.com
Make It Your Own: Autotelic Curiosity, with Pierre-Yves Oudeyer
Research Pierre-Yves Oudeyer and his team are studying autotelic curiosity, that marvelous ability to self-generate and self-select our own learning objectives, goals or games. They design robots to l...
060
Reposted by Pierre-Yves Oudeyer
Basque Center on Cognition, Brain and Language (BCBL) @bcbl.bsky.social · 13/04/2026
[2/2] Joining us will be @jennysaffran.bsky.social, from @uwmadison.bsky.social; @noranewcombe.bsky.social, from Temple University; @pyoudeyer.bsky.social, from Inria; and Noam Sigelman, from The Hebrew University of Jerusalem. 📅 June 10-12, 2026 +info ⬇️ www.bcbl.eu/events/stati...
031
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
⚠️ Small sample — precise results should be taken with caution. But we've already reproduced most findings in another study we'll share soon. 📄 Paper: arxiv.org/abs/2505.01106
arxiv.org
The Illusion of Understanding: How Middle-Schoolers Fail to Regulate Inquiry with ChatGPT in a Science Task
Generative AI (GenAI) tools allow for effortless task completion, potentially fostering cognitive and metacognitive laziness in students. While surveys indicate widespread GenAI use among students as ...
0122
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
🎯 Takeaways for educators: → AI literacy must train question-asking & answer evaluation, not just "how AI works" → Training metacognition should be a priority: benefits far beyond AI use → An "effortless" AI interaction should be a red flag, not a comfort signal
2186
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
But not all students were equal: 🧠 Those with stronger metacognitive skills (monitoring & regulating their own learning) judged prompt quality better ⚠️ Paradoxically, self-rated AI-experienced students performed worse: self-taught AI breeds overconfidence, not critical skills
191
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
3️⃣ They almost never asked follow-up questions — even after incomplete answers. Only 14 out of 63 students ever did.
160
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
2️⃣ They couldn't tell a good AI answer from a poor one. Vague responses were rated "useful" just as often as genuinely informative ones. Prior knowledge didn't help: confident students were just as fooled by fluent-sounding text.
170
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
1️⃣ Students couldn't distinguish a good prompt from a bad one. A precise, context-rich question vs. a vague one? Chance-level discrimination.
150
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
So what went wrong? In short: cognitive surrender. ChatGPT's fluent, confident tone creates an illusion of understanding. This manifests in 3 concrete ways 👇
172
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
We studied 63 students (aged 14-15) in French schools. Task: read a science inquiry problem, query ChatGPT, evaluate its answers, then explain the concept in your own words. 📊 Average score: 10/20 — despite unlimited AI access and solvable problems (some scored near-perfect).
131
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 01/04/2026
🔍 What happens when you give middle-schoolers unrestricted access to ChatGPT for science tasks? Findings from our new paper led by Rania Abdelghani w/ @koumurayama.bsky.social @celestekidd.bsky.social Hélène Sauzéon 🧵👇
24418
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
To learn more about the extraordinary PhD work of Nicolas Yax: www.linkedin.com/posts/pierre...
linkedin.com
Are we too human to understand LLMs? 🤔 This question threads through Nicolas Yax's outstanding PhD 💡, which he brilliantly defended a few days ago (co-supervised with the great Stefano Palminteri...
Are we too human to understand LLMs? 🤔 This question threads through Nicolas Yax's outstanding PhD 💡, which he brilliantly defended a few days ago (co-supervised with the great Stefano Palminteri) H...
110
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
📖 Blog: developmentalsystems.org/phylolm 📄 ICLR 2025 paper: iclr.cc/virtual/2025... 💻 Code: github.com/Nicolas-Yax/... Hugging Face online demo: t.co/9wEdav3LZA
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
This work was led by the outstanding @nicolasyax.bsky.social and co-supervised by @stepalminteri.bsky.social and me
110
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
The blog walks the full intellectual journey — from cat genetics to Dawkins' memes, from myth phylogenetics to the evolutionary landscape of modern AI. Written to be accessible to anyone curious about evolution, culture, and intelligence. 🌱
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
Why does this matter? 300+ models appear daily on @huggingface . Training transparency is declining. Tools like PhyloLM could matter for AI governance, safety monitoring, and simply mapping the evolutionary landscape of what's out there.
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
Result: surprisingly accurate reconstruction of known model family trees. And a few insights on the ancestry of closed-source models 👀
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
Our method PhyloLM (ICLR 2025) treats language models as populations of text. "Genes" = short prompt contexts. "Alleles" = their completions. Apply classic genetic distance measures → build phylogenetic trees. No prior knowledge of training history needed.
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
A good evolutionary marker needs 3 properties: → compressed & universal representation → moderate rate of change → functional grounding DNA has all three. So do high-level myth features. And so does an LLM's core identity: its probability distribution over text.
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
Now the key question: can you apply the same method to LLMs? Without access to their code, weights, or training data? Mostly Yes. 🧬
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
The myth phylogenies match known human migration patterns — shown beautifully by the outstanding work of Julien d'Huy. Same logic, different substrate: not DNA, but narrative structure as evolutionary marker.
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
Can you reconstruct the evolutionary tree of cat breeds just by looking at their genes? Yes — biologists do this routinely. Can you do the same with ancient myths, tracing how stories evolved across continents by comparing narrative structure? To a large extent, yes.
100
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/03/2026
New blog post: The Phylogenetics of Artifacts — inferring the evolution of cultural objects, artificial life forms, and language models. From cat genetics to ancient myths to LLMs. 🧬 1/n
3167
Reposted by Pierre-Yves Oudeyer
Joanna Bryson @j2bryson.bsky.social · 10/03/2026
Wow, some of my old #EvolutionOfLanguage #EoL pals may have just done something huge for #AILaw data attribution #LLM #AIGovernance Specifically @pyoudeyer.bsky.social @nicolasyax.bsky.social @stepalminteri.bsky.social Evolutionary biology can track LLM phylogeny! developmentalsystems.org/phylolm
developmentalsystems.org
The Phylogenetics of Artifacts - A deep dive into the evolution of cultural objects, artificial life forms and language models
Developmental Systems, a Blog of the Flowers Lab
185
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 16/07/2025
Major new system and result from the team: SOAR is an open-source self-improving genAI system pushing the frontier of ARC performances using program synthesis It relies on using LLMs as self-improving smart operators for evolutionary search
140
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 16/07/2025
Congratulations @jul-p.bsky.social for this major achievement and @ccolas.bsky.social for the amazing co-supervision ! Your work was magic to develop this self-improving system pushing the frontier of what can be done with program synthesis and open-source methods and models on the ARC challenge !
020
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/06/2025
Using LLMs to advance the cognitive science of collectives Very interesting new paper by @sucholutsky.bsky.social Katherine Collins @norijacoby.bsky.social @billdthompson @roberthawkins.bsky.social arxiv.org/pdf/2506.00052
030
Reposted by Pierre-Yves Oudeyer
Antoine Blanchard @enroweb.com · 06/06/2025
Le travail de Samuel Bianchini à partir des algorithmes de @pyoudeyer.bsky.social mis en lumière par le Jeu de paume dans son expo IA
Catalogue de l’expo « Le monde selon l’IA »Artiste et enseignant-chercheur, Samuel Bianchini collabore avec de nombreux ingénieurs et scientifiques pour explorer les relations qu'entretiennent les nouvelles technologies avec les contextes culturels dans lesquels elles s'insèrent.
Dans cette troisième version de l'installation Prendre vie(s), Samuel Bianchini mobilise Flow Lenia, un logiciel de vie artificielle augmentee par IA. Cette forme d'animation, née d'une simulation mathématique appelée «jeu de la vie » en 1970, compose des systèmes susceptibles d'adopter des comportements émergents et inattendus.

Prendre vie(s), version 03, 2020-2025
Développement informatique (algorithmes de vie et d'intelligence artificielles):
Léon Denise et Adrian Mangel, sur la base de l'environnement logiciel Flow Lenia développé par l'équipe Flowers (Inris.
Université de Bordeaux - Pierre-Yves
Oudeyer, Clément Moulin-Fricr, Gautier Hamon et Erwan Plantee) à partir de Lenis. développé par Bert Chan (DeepMind). et avec la collaboration de Colin Bouvry Ce projet a bénéficié du soutien de l'École des arts décoratifs (Université PSL, Paris) et du festival accès)s( cultures électroniques
Remerciements: Alain Declercq.
Jean-Jacques Gay. Stéphane Trois Carrés
131
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/06/2025
Merci pour le lien ! Pour être plus précis, ce sont les algorithmes développés par des membres de mon équipe, ici en particulier @eplantec.bsky.social @clemmoulinfrier.bsky.social @hamongautier.bsky.social et le système Flow Lenia sites.google.com/view/flowlen...
sites.google.com
FLOW LENIA
Erwan Plantec, Gautier Hamon, Mayalen Etcheverry, Pierre-Yves Oudeyer, Clément Moulin-Frier, Bert Wang-Chak Chan Inria Flowers and Google Deepmind
030
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 28/05/2025
guydavidson.me/files/goal_i...
guydavidson.me
050
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 28/05/2025
Humans' ability to invent their own games & goals is at the core of open-ended learning. Understanding and modeling computationally how they do it would be enlightening to understand better human cognition and build open-ended AI Great step in this direction in new paper by Guy Davidson et al.
1194
Reposted by Pierre-Yves Oudeyer
Melanie Mitchell @melaniemitchell.bsky.social · 19/05/2025
I reviewed "These Strange New Minds: How AI Learned to Talk and What It Means" by Chris Summerfield. melaniemitchell.me/EssaysConten...
14611
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 15/05/2025
royalsocietypublishing.org/toc/rstb/202...
020
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 15/05/2025
Cool
040
Reposted by Pierre-Yves Oudeyer
Marko Cvjetko @mcvjetko.bsky.social · 09/05/2025
Another one.
041
Reposted by Pierre-Yves Oudeyer
Marko Cvjetko @mcvjetko.bsky.social · 09/05/2025
Here is a cool Flow-Lenia simulation, which is a continuous cellular automaton with two distinct features: 1. Mass conservation - sum of all activations is constant through time 2. Localized update rules - enables simulations with differently behaving matter on the same grid
1209
Reposted by Pierre-Yves Oudeyer
Clément Moulin-Frier @clemmoulinfrier.bsky.social · 12/05/2025
Very happy that our paper received both the EvoApps best paper and the EvoStar best student paper! Congrats team!! Emergent Kin Selection of Altruistic Feeding via Non-Episodic Neuroevolution arxiv.org/abs/2411.10536
arxiv.org
Emergent kin selection of altruistic feeding via non-episodic neuroevolution
Kin selection theory has proven to be a popular and widely accepted account of how altruistic behaviour can evolve under natural selection. Hamilton's rule, first published in 1964, has since been exp...
0104
Reposted by Pierre-Yves Oudeyer
Stefano Palminteri @stepalminteri.bsky.social · 29/04/2025
Check out @nicolasyax.bsky.social thread about our paper (co-supervised by @pyoudeyer.bsky.social) where we show that evolutionary tree reconstruction can be successfully applied to map LLMs to map relations and predict their performance! Currently at @iclr-conf.bsky.social
041
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 25/04/2025
arxiv.org/abs/2407.04503
arxiv.org
When LLMs Play the Telephone Game: Cumulative Changes and Attractors in Iterated Cultural Transmissions
As large language models (LLMs) start interacting with each other and generating an increasing amount of text online, it becomes crucial to better understand how information is transformed as it passe...
000
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 25/04/2025
And thank you @gaiamolinaro.bsky.social for presenting the work ! + thanks Jérémy Perez for leading the project and thanks the whole team Corentin Leger @kovacgrgur.bsky.social @ccolas.bsky.social @clemmoulinfrier.bsky.social @maximederex.bsky.social
110
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 25/04/2025
Generative AI is a cultural transmission technology: it plays a growing role in generation, selection and transmission of ideas/opinions in human society 🧠🔄🌐 And yet we understand very little of this dynamics at this point 🤔❓ A step forward is our #ICLR2025 paper !👇
171
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 24/04/2025
@thomwolf.bsky.social This is the HF space we discussed, allowing to map the space of all models on HF, and infer which derive from where :)
000
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 24/04/2025
Congratulations @nicolasyax.bsky.social for the amazing work, and @stepalminteri.bsky.social for the co-supervision !
021
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 24/04/2025
I have to say that when Nicolas pitched us the idea of this project, I was excited about the creativity, but was not convinced it would work: I was completely wrong, it worked close to directly! Beautiful!
000
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 24/04/2025
Imagine you can compute a (very) cheap behavioural signatures for LLMs and from it 1) infer their history (which previous models they derived from) 2) build large maps to navigate spaces of 100s of models 3) predict (coarsely) their performances in benchmarks That's PhyloLM 🚀
010
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 16/04/2025
We are starting a new major research project investigating the joint development of curiosity and metacognition in adolescents, where the Flowers lab will collaborate with @mjgruber.bsky.social 's lab at Univ. Cardiff, and Yana Fandakova at Univ. Trier !!! Open postdocs positions, see below 👇
061
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 10/04/2025
Why Self-Determination Theory Needs Computational Modelling: The Case of Competence and Optimal Challenge by @codingconduct.cc E. Lintunen N. Aly @creativeendvs.bsky.social osf.io/n6x8s/downlo...
021
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 03/04/2025
Thanks @ccolas.bsky.social for pointing it out ! This has great potential for diversity search algorithms
010
Pierre-Yves Oudeyer @pyoudeyer.bsky.social · 03/04/2025
Forcing Diffuse Distributions out of Language Models "When LLMs are used for real-world tasks where diversity of outputs is crucial their inability to produce diffuse distributions over valid choices is a major hurdle" New method to address this challenge by Zhang et al.: arxiv.org/pdf/2404.10859
131
Reposted by Pierre-Yves Oudeyer
Konrad Kording @kordinglab.bsky.social · 26/03/2025
Accepting “the bitter lesson” and embracing the brain’s complexity by Dyer and Richards. Great article! www.thetransmitter.org/neuroai/acce...
thetransmitter.org
» Accepting “the bitter lesson” and embracing the brain’s complexity Close menuMenuClose menuSpectrum logoYouTubeBlueskyMastodonYouTubeBlueskyMastodon
3339