Fetching the paper…

IDA-VLM: Towards Movie Understanding via ID-Aware Large Vision-Language Model · Around