Mitigating Hallucination in Large Vision-Language Models through Aligning Attention Distribution to Information Flow.

Jianfei Zhao, Feng Zhang, Xin Sun, Chong Feng

Browse the full EMNLP paper archive.