Konstantinos Voudouris Marta Halina Murray Shanahan Joel Holmes 🇬🇧🇱🇻🇺🇦🇵🇸 Wout Schellaert John Burden Don't forget Benjamin Beyret implemented the first version from scratch and shaped what was possible to build with it.

New paper: We ask How does changing the context of questions affect whether LLMs act rationally? We find that popular LLMs are extremely sensitive to the question context. We show this with thousands of variations of the Prisoner's Dillemma wrapped up in different contexts.

But today, linguistics sits at the crossroads of numerous disciplines, including biology, law and computer science. Colleagues from across academia are suddenly interested in leveraging what linguisticians have learned. news.berkeley.edu/2025/03/13/wit…

🚨 We're hiring! If you're excited about 🤖 ML/LLMs, 🧠 cognitive science, or 💭 computational psychiatry, come join us in Munich. Two fully funded PhDs Helmholtz Munich | @HelmholtzMunich: great mentorship, international vibe, lots of room to grow. 📅 May 16th 🔗 hcai-munich.com/PhDHCAI.pdf

Excited to say our paper got accepted to ICML! We added new findings including this: models fine-tuned on a visual counterfactual reasoning task do not generalize to the underlying factual physical reasoning task, even with test images matched to the fine-tuning data set.

LLMs • agentic AI • #DataScience 🧵 1/ 🚨 New paper: “Measuring Data-Science Automation: A Survey of Evaluation Tools for AI Assistants & Agents.” If you care about the impact of LLMs and LLM agents on Data Science and how to measure it, this is for you!

Excited to see our Centaur project out in nature. TL;DR: Centaur is a computational model that predicts and simulates human behavior for any experiment described in natural language.