📣9th Summer School on Computational Interaction in June in Paris🇫🇷. Speakers include Lydia Chilton, John Williamson, Laure Soulier, Jean Claude Dreher and myself. Application deadline March 14th. cixschool2025.isir.upmc.fr

Andy Clark is one of the leading thinkers in human cognition and our emerging cyborg future. He has written some of the most influential work on extended mind theory and human-technology integration! nature.com/articles/s4146…

Is RL really scalable like other objectives? We found that just scaling up data and compute is *not* enough to enable RL to solve complex tasks. The culprit is the horizon. Paper: arxiv.org/abs/2506.04168 Thread ↓

🧵 1/8 The Illusion of Thinking: Are reasoning models like o1/o3, DeepSeek-R1, and Claude 3.7 Sonnet really "thinking"? 🤔 Or are they just throwing more compute towards pattern matching? The new Large Reasoning Models (LRMs) show promising gains on math and coding benchmarks,

Overall, this paper gives a better understanding of how models learn and how they memorize Paper: arxiv.org/abs/2505.24832

Our group meeting today is in Aalto'a cinema theater. No 🍿 though

What if LLMs could learn your habits and preferences well enough (across any context!) to anticipate your needs? In a new paper, we present the General User Model (GUM): a model of you built from just your everyday computer use. 🧵

Is AI better than doctors at diagnosing patients? A new study led by researchers at Stanford University Stanford Medicine and Eric Horvitz Microsoft answered this question with gold-standard evidence: a randomized controlled trial (RCT). 👇 70 clinicians (residents + attendings) were

Philipp Wintersberger Philipp Wintersberger giving a talk on multitasking and human-AI interaction Aalto University

John Williamson lecturing about simulator-bases inference at the Comp Int summer school cixschool2025.isir.upmc.fr

John Williamson lecturing about simulator-bases inference at the Comp Int summer school cixschool2025.isir.upmc.fr

We had a lovely summer school last week in Paris. Thirty amazing students and postdocs, a week full of keynotes and hands-on lectures on computational methods. State-of-the-art as call as classical topics, ranging from LLM agents to Bayesian inference. cixschool2025.isir.upmc.fr

Microsoft claims their new AI framework diagnoses 4x better than doctors. I'm a medical doctor and I actually read the paper. Here's my perspective on why this is both impressive AND misleading ... 🧵