Increasing the Difficulty of Automatically Generated Questions via Reinforcement Learning with Synthetic Preference for Cost-Effective Cultural Heritage Dataset Generation

Increasing the Difficulty of Automatically Generated Questions via Reinforcement Learning with Synthetic Preference for Cost-Effective Cultural Heritage Dataset Generation William Thorne author Ambrose Robinson author Bohua Peng author Chenghua Lin author Diana Maynard author 2024-11 text Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities Mika Hämäläinen editor Emily Öhman editor So Miyagawa editor Khalid Alnajjar editor Yuri Bizzoni editor Association for Computational Linguistics Miami, USA conference publication thorne-etal-2024-increasing 10.18653/v1/2024.nlp4dh-1.43 https://aclanthology.org/2024.nlp4dh-1.43/ 2024-11 450 462