Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection

Open in new window