This video shows my Tetris test of Laya. Laya is an open-source decision model that uses BERT-style encoders to answer structured questions in a single forward pass. Its API supports selecting from a set of options and returning numerical scores or probabilities.

Download the video (.mov).

Laya reached 92.33% accuracy in my IMDb movie review classification test and failed the Tetris test shown here. I discuss these results and the broader classification context in Language Models for Text Classification: From Bag-of-Words to Jev.