Voila-A: Aligning Vision-Language Models with User's Gaze Attention

In recent years, the integration of vision and language understanding has led to significant advancements in artificial intelligence, particularly through Vision-Language Models (VLMs).