🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

I've now been in research long enough to Schmidhuber Schmiduber!! Life goal achieved. ✅

Really excited to share our recent work combining open-ended foundation model innovation with the compeititive dynamics of self-play!! One of the ingredients towards creativity explosion? Led by Aaron Dharna, done together with Jeff Clune 🚀

we're introducing a new self-improving evolutionary search framework and apply it to get good performance on ARC-1! simple idea: search with llms, hindsight relabel solutions, finetune the llm, and repeat can be done with or without ground truth labels (test time training)

Can an AI model predict perfectly and still have a terrible world model? What would that even mean? Our new ICML paper formalizes these questions One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵

This but with agents

OMW to #ICML2025, reach out (on X or Whova) if you’re interested in talking about RL, reasoning/safety of LLMs, agents. I will also be presenting our AI4MATH workshop paper: limits-of-RLVR, 13:45-14:00 pm, July 18, 2025. Location: Ballroom C, feel free to drop by

Program Synthesis approach breakthrough in ARC-AGI through Self-Play 📖 Read 201: « Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on Arc-AGI », by Pourcel Julien @ICML, Cédric and Pierre-Yves Oudeyer github.com/flowersteam/SO… The work of the authors is a

✨Introducing SENSEI✨ We bring semantically meaningful exploration to model-based RL using VLMs. With intrinsic rewards for novel yet useful behaviors, SENSEI showcases strong exploration in MiniHack, Pokémon Red & Robodesk. Accepted at ICML 2025🎉 Joint work with Christian Gumbsch 🧵

Nice paper that in my opinion goes in the right direction to solve ARC. It generates python code to tackle the ARC tasks and combines search and learning in a virtuous cycle. I have summarized the results in the following plot.

I'm attending ICML 2025 this week in Vancouver where we're presenting our MAGELLAN paper along with Loris Gaven and Carta Thomas! 📅 Come discuss at our poster session on July 17 at 11 am East Exhibition Hall A-B E-2803 Or reach out for a chat! x.com/CartaThomas2/s…

Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI by Pourcel Julien @ICML, Cédric and Pierre-Yves Oudeyer Another example of ARC-AGI as a research playground that has general applicability

A promising approach for AGI: "SOAR" 52% on ARC-AGI in one big jump, using program synthesis, evolutionary search, and learning. Definitely worth keeping an eye on.

Want to find out more about SOAR? Come chat with us at ICML and learn more! 📍 East Exhibition Hall A-B, Booth E-2807 🗓️ Tuesday, 15 July ⏰ 4:30 – 7:00 p.m. PDT Cédric Pierre-Yves Oudeyer

I’m attending #ICML this week! We’ll be presenting MAGELLAN during the poster session on Thursday with Carta Thomas & Clément ROMAC @ ICML 2025 If you’re not in Vancouver, we recorded a talk presenting the paper last week, it’s available on YouTube (link below)

Wow. 52% on ARC-AGI 1 just from a few %, using program synthesis, evolutionary search, and learning. Another self improvement startup starts!

Announcing Mistral AI Voxtral, our frontier open source speech understanding models.