作者位次 #4/6ACL 2026Findings· Findings of ACL
V is TW : Benchmarking Vision-Language Models for T aiwanese M andarin in T aiwan
Zhi Rui Tam · Yung-Yu Shih · Yen-Wei Lee · Ya-Ting Pai · Wen Yu Chang · Yun-Nung Chen
摘要
Vision-Language Models (VLMs) often struggle in Taiwanese Mandarin environments due to region-specific orthographic and cultural context. We introduce VisTW, a comprehensive benchmark featuring (i) multiple-choice questions (3,795 academic questions) and (ii) free-form generation evaluation (141...