Search for: All records

In this post I want to talk about using generative AI to extend one of my academic software projects—the Python Tutor tool for learning programming—with an AI chat tutor. We often hear about GenAI being used in large-scale commercial settings, but we don’t hear nearly as much about smaller-scale not-for-profit projects. Thus, this post serves as a case study of adding generative AI into a personal project where I didn’t have much time, resources, or expertise at my disposal. Working on this project got me really excited about being here at this moment right as powerful GenAI tools are starting to become more accessible to nonexperts like myself. 

We present Seq2Parse, a language-agnostic neurosymbolic approach to automatically repairing parse errors. Seq2Parse is based on the insight that Symbolic Error Correcting (EC) Parsers can, in principle, synthesize repairs, but, in practice, are overwhelmed by the many error-correction rules that are not relevant to the particular program that requires repair. In contrast, Neural approaches are fooled by the large space of possible sequence level edits, but can precisely pinpoint the set of EC-rules that are relevant to a particular program. We show how to combine their complementary strengths by using neural methods to train a sequence classifier that predicts the small set of relevant EC-rules for an ill-parsed program, after which, the symbolic EC-parsing algorithm can make short work of generating useful repairs. We train and evaluate our method on a dataset of 1,100,000 Python programs, and show that Seq2Parse is accurate and efficient : it can parse 94% of our tests within 2.1 seconds, while generating the exact user fix in 1 out 3 of the cases; and useful : humans perceive both Seq2Parse-generated error locations and repairs to be almost as good as human-generated ones in a statistically-significant manner.