SyntaxGym is a unified platform where language and NLP researchers can design psycholinguistic tests and visualize the performance of language models. Our goal is to make psycholinguistic assessment of language models more standardized, reproducible, and accessible to a wide variety of researchers. The project is run out of the MIT Computational Psycholinguistics Laboratory.
Interested in viewing or designing psycholinguistic test suites? Create a new test suite online or upload one as a
Have a model you want to evaluate? Add a model as a Docker container, and it will automatically be evaluated on existing test suites.