Published in the Proceedings of the Tenth Italian Conference on Computational Linguistics (CLiC-it 2024), Pisa, Italy, CEUR Workshop Proceedings.
The paper builds a tool that generates Italian crossword puzzles from educational text using GPT-4o, Mistral-7B-Instruct-v0.3 and Llama3-8B-Instruct, supported by Italian-Clue-Instruct, a dataset of over 30,000 entries. It covers four clue styles: unconstrained clues, definite determiner phrases, copular sentences and bare noun phrases.
Authors: Kamyar Zeinalipour, Achille Fusco, Asya Zanollo, Marco Maggini, Marco Gori.
Read the paper | Code | Models and data
All the lab's language releases are listed in Language models and datasets.