Abulhair Saparov

prontoqa. Synthetic question-answering dataset to formally analyze the chain-of-thought output of large language models on a reasoning task.

parser. Semantic parser induction using a generative model of grammar.

PWL. Natural language understanding by probabilistic abduction of a symbolic theory from sentences and logical forms.

grammar. Implementation of generative semantic grammar.

learning_to_search. Train transformers on synthetic graph search problems to measure their scaling behavior and to do mechanistic analysis.

fictionalgeoqa. Question-answering dataset to evaluate reasoning ability over short paragraphs.

core. Small platform-independent template library with general-purpose data structures and algorithms.

hdp. Code to work with Dirichlet processes and hierarchical Dirichlet processes and perform inference using MCMC.

archivist. Python script to convert Doxygen XML into static HTML webpages for C++ documentation.

math. Small collection of data structures for performing common mathematical operations in scientific computing.