Identifying and Mitigating Position Bias of Multi-image Vision-Language Models

Re-assign community