Investigating and Mitigating Object Hallucinations in Pretrained Vision-Language (CLIP) Models.

Yufang Liu, Tao Ji, Changzhi Sun, Yuanbin Wu, Aimin Zhou

Browse the full EMNLP paper archive.