作者位次 #3/5ACL 2026Findings· Findings of ACL
Once Correct, Still Wrong: Counterfactual Hallucination in Multilingual Vision-Language Models
Basel Mousi · Fahim Dalvi · Shammur Absar Chowdhury · Firoj Alam · Nadir Durrani
摘要
Vision–language models (VLMs) can achieve high accuracy while still accepting culturally plausible but visually incorrect interpretations. Existing hallucination benchmarks rarely test this failure mode, particularly outside Western contexts and English. We introduce M ^2 CQA , a culturally groun...