Presented at the International Workshop on AI in Education and Educational Research (AIEER 2024), part of the 27th European Conference on Artificial Intelligence (ECAI 2024), Santiago de Compostela, Spain.
The paper introduces Essay-Syntax-Instruct, a dataset for helping students find and fix English syntax errors, and fine-tunes GPT-3.5-Turbo, Llama-2-7B-chat, Llama-2-13B-chat and Mistral-7B-Instruct on it. The fine-tuned models showed clear improvements on syntax correction.
Authors: Kamyar Zeinalipour, Mehak Mehak, Fatemeh Parsamotamed, Marco Maggini, Marco Gori.
Read the paper | Code | Models and data
All the lab's language releases are listed in Language models and datasets.