Researchers at the Georgia Institute of Technology have created a dataset of conversations about recipes to improve language models used for cooking. The dataset, named ChattyChef, contains 267 dialogues between one person who does not know the recipe steps and another who has access to the recipe and culinary expertise. Researchers can use the dataset to train language models to better understand users’ intent during tasks with multiple steps and respond accordingly.
Image credit: Flickr user Floris van Halm
