Description
The rapid advancement of Generative Artificial Intelligence and Large Language Models (LLMs) has significantly transformed the landscape of Natural Language Processing (NLP), particularly in semantic understanding and multilingual language processing. This study investigates the application of LLMs to English–Vietnamese bilingual semantic role labeling, with an emphasis on enhancing semantic representation for a low-resource language. Specifically, the study explores the effectiveness of transfer learning and fine-tuning strategies in improving semantic role identification and evaluating the semantic reasoning capabilities of LLMs in bilingual contexts. Beyond empirical performance, the paper examines the social responsibility implications of deploying generative AI, including semantic bias, linguistic fairness, and transparency in language processing. The findings are expected to contribute to the development of more reliable, interpretable, and socially responsible bilingual NLP systems, while providing practical insights into the application of LLMs for Vietnamese computational linguistics.
Từ khóa
Generative Artificial Intelligence; Large Language Models; Semantic Role Labeling; Low-Resource Languages; Responsible AI
Thông tin các tác giả
1/ Huỳnh Quang Đức, ThS, đang công tác tại Trường Đại học Bình Dương. Số 504 Đại Lộ Bình Dương, Phường Phú Lợi, Tp. Hồ Chí Minh. Email: hqduc@bdu.edu.vn
2/ Nguyễn Hồ Hải, ThS, đang công tác tại Trường Đại học Bình Dương. Số 504 Đại Lộ Bình Dương, Phường Phú Lợi, Tp. Hồ Chí Minh. Email: nhhai@bdu.edu.vn
