作者位次 #4/4ACL 2026Poster· Natural Language Processing
A ra VQA : Building a New A rabic Factoid Visual Question Answering Dataset from W ikipedia
Sultan Alrowili · Younes Samih · Abed Alhakim Freihat · Mathan Kumar Eswaran
摘要
The development of large-scale Visual Question Answering (VQA) datasets has traditionally relied on resource-intensive manual annotation. In addition, most of the existing Arabic VQA datasets focus on culturally-specific and dialect-aware domains. To address these limitations, we propose a new pi...