Mitigating Object Hallucinations in Large Vision-Language Models via Attention Calibration

Open in new window