Testing the limits of natural language models for predicting human language judgments

1 code implementation • 7 Apr 2022 • Tal Golan, Matthew Siegelman, Nikolaus Kriegeskorte, Christopher Baldassano

Neural network language models can serve as computational hypotheses about how humans process language.

Sentence

Learning to Perform Role-Filler Binding with Schematic Knowledge

1 code implementation • 24 Feb 2019 • Catherine Chen, Qihong Lu, Andre Beukers, Christopher Baldassano, Kenneth A. Norman

We can perform these bindings for arbitrary fillers -- we understand this sentence even if we have never heard the names "Alice," "tea," or "Bob" before.

Question Answering Sentence

Mapping Between fMRI Responses to Movies and their Natural Language Annotations

1 code implementation • 13 Oct 2016 • Kiran Vodrahalli, Po-Hsuan Chen, YIngyu Liang, Christopher Baldassano, Janice Chen, Esther Yong, Christopher Honey, Uri Hasson, Peter Ramadge, Ken Norman, Sanjeev Arora

Several research groups have shown how to correlate fMRI responses to the meanings of presented stimuli.

Scene Classification Sentence +2

Affordances Provide a Fundamental Categorization Principle for Visual Scenes

no code implementations • 19 Nov 2014 • Michelle R. Greene, Christopher Baldassano, Andre Esteva, Diane M. Beck, Li Fei-Fei

Traditional models of visual perception posit that scene categorization is achieved through the recognition of a scene's objects, yet these models cannot account for the mounting evidence that human observers are relatively insensitive to the local details in an image.