Large Language Models Engineer Too Many Simple Features For Tabular Data

Welcome to the LLMs Feature Engineering Bias repository!

This project is still in development, and there are no active files regarding bias evaluation at the moment. This repository contains the implementations from the paper. The proposed context-aware feature generation method, as defined in the paper, is publicly available at the CAAFG repository.