3 papers
cs.CL2026
Fodor and Pylyshyn's Systematicity Challenge Still Stands
Michael Goodale, Salvador Mascarenhas
The recent successes of neural networks producing human-like language have caused significant stir in cognitive science, with many researchers arguing that classical puzzles about…
cs.LG2025
MapFormer: Self-Supervised Learning of Cognitive Maps with Input-Dependent Positional Embeddings
Victor Rambaud, Salvador Mascarenhas, Yair Lakretz
A cognitive map is an internal model which encodes the abstract relationships among entities in the world, giving humans and animals the flexibility to adapt to new situations, wit…
cs.CL2025
Meta-Learning Neural Mechanisms rather than Bayesian Priors
Michael Goodale, Salvador Mascarenhas, Yair Lakretz
Children acquire language despite being exposed to several orders of magnitude less data than large language models require. Meta-learning has been proposed as a way to integrate h…