After four incredible years in the Flowers AI & CogSci Lab, I’m thrilled to share that I’ve officially become a Doctor! 🎓💫 My thesis: “Language as a Cognitive Tool for Open Agents.” It explores how language helps artificial agents explore, adapt, and learn efficiently.

I was so excited when I saw that Neural Cellular Automata had been applied to the @arcprize - do you remember the famous self-repairing Gecko from @zzznah in 2020? I think ALIFE methods are an exciting path forward in AI research! This is @stenichele and Etienne Guichard at

3 papers from @FlowersINRIA are presented this week at the @ALifeConf , leveraging curiosity-driven AI to explore complex behaviors in #Lenia All first authors are attending the conference if you want to discuss with them Link to interactive websites, papers and code below:

A butterfly is born 🦋 This is a new type of CA called "MaCE Lenia", by @VassPapas & E Guichard. arXiv paper: https://t.co/uYURjFJYdM They'll have a poster & demo at @ALifeConf, (I will also have a booth nearby). Please come to visit us at Posters + Public Expo (Tue 5-8pm)!

Lenia forest fire model

Introducing SOAR 🚀, a self-improving framework for prog synth that alternates between search and learning (accepted to #ICML!) It brings LLMs from just a few percent on ARC-AGI-1 up to 52% We’re releasing the finetuned LLMs, a dataset of 5M generated programs and the code. 🧵

Synthetic ecosystems that autonomously and continuously evolve in silico 👾 An Alifer dream we pursue with Flow-Lenia ! If you are interested in complex systems with (1) emergent creatures and (2) intrinsic evolutionary dynamics, go check our new paper ! A 🧵

🚀 New preprint! ARC-NCA shows how Neural Cellular Automata —incl. memory-rich EngramNCA—crack tasks from the ARC-AGI benchmark, hitting GPT-4.5-level accuracy at a fraction of the cost. 🌱🤖 Web: https://t.co/e5iWq1jf16 Preprint: https://t.co/ltm4RPM9Zy #NeuroAI #AGI #CA #ALIFE

Here are two cool simulations of Flow-Lenia, 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

New paper! @BeneHartl and @AndreasZoettl https://t.co/GbtvAFcErU "Neuroevolution of decentralized decision- making in N-bead swimmers leads to scalable and robust collective locomotion" "Many microorganisms swim by performing larger non-reciprocal shape deformations that are

Ah, all the artifacts were fixed by replacing one '<' with '<=', duh. It all works now, particle life completely on WebGPU! There are 2^18 (~260k) particles here; my PC struggles after about half a million particles. Will probably try to optimize it a bit later.

i just got an art grant from the council for the arts at MIT! *Tangible Dreams* will let visitors experiment and play with a physical neural network generating images real-time—by twisting knobs and switches, by reconnecting nodes together @ArtsatMIT

Time-lapse of cultured cells crawling and exploring. A nebulous galaxy not in outer space but in our inner space.

> GPT model stopped speaking Croatian > Nobody could figure out why. Turns out > Croatian users were much more prone downvote messages

Ever wonder what the architecture of a neural network would look like, in a novel organism that had not been through selection for specific structure and function of an embodied nervous system? Here's our #preprint with morphological, behavioral, electrophysiological, and

Starting a PhD is associated with a considerable worsening of mental health. After entering their programs, students start seeking *a lot* more psychiatric medication for managing that bad mental health.

👀

🧠 One of the key limitation of LLMs today is their lack of metacognition: they were (mostly) not trained to know what they know or don't know, what they can or can't do. 🚀At @FlowersINRIA, we're proposing an approach to build metacognition into LLMs: MAGELLAN !

Complex cell-like structures in Flow Lenia (continuous cellular automaton with built-in mass conservation)

Announcing SONI: Self-Organising Neural Intelligence(s) 🧠 An #ALife2025 special session to bring together AI & @ALifeConf communities to explore how self-organisation can contribute to neural paradigms of intelligence Call for papers https://t.co/I5IAFMfXe9 |Deadline: May 4th