A arxiv.org Benchmarking vision-language models on OCR in dynamic video environments 142 points ashu_trv 1 year ago 58 comments
