Tyler Brooke-Wilson
@T_BrookeWilson
Incoming asst prof @Yale. Thinking about how to computationally model the mind. Previously @MIT.
How do people reason while still staying coherent – as if they have an internal ‘world model’ for situations they’ve never encountered? A new paper on open-world cognition (preview at the world models workshop at #ICML2025!)
Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models
One of the most exciting papers I've read in a while - highly recommended!🧠🤖 It gives a compelling account of how the human mind reasons so flexibly, and of what's missing from LLM reasoning
How do people reason while still staying coherent – as if they have an internal ‘world model’ for situations they’ve never encountered? A new paper on open-world cognition (preview at the world models workshop at #ICML2025!)
I think this also suggests interesting directions for studying the limits of open-ended cognition in humans & AI -- ad-hoc model synthesis is a hard problem, so how is it possible at all? (cont.) x.com/T_BrookeWilson…
Our hypothesis: people construct ad-hoc world models on the fly, in response to task demands, with just the relevant variables needed to support reasoning in a given situation. We call this a ‘Model Synthesis Architecture’:
I wonder where the log-x scale comes from - is that an inherent property of the tasks being evaluated on, or a sign of some limitation in models' search/reasoning strategy?