BIMA: Bijective Maximum Likelihood Learning Approach to Hallucination Prediction and Mitigation in Large Vision-Language Models.

Huu-Thien Tran, Thanh-Dat Truong, Khoa Luu

Browse the full CVPR paper archive.