Researchers have extended a visual document retrieval (VDR) system to support multiple languages. The system, called VDR-2B, uses computer vision and natural language processing techniques to search for documents based on images or text queries. It can now handle queries in up to 100 languages, including those with non-Latin scripts such as Chinese, Japanese, and Korean. --- Why it matters: This development matters because it enables the use of visual document retrieval systems across languages and cultures, potentially improving access to information for people who speak diverse languages. Source: https://huggingface.co/blog/vdr-2b-multilingual