Show GN: VLM은 한국 공공기관 문서를 얼마나 잘 읽을까? KOLongDoc 벤치마크 공개

As multimodal AI models like ChatGPT and Claude increasingly support Korean public administration, KOLongDoc fills a critical gap by providing the first comprehensive benchmark for evaluating how well VLMs understand long Korean government documents. This enables organizations to accurately assess model suitability for real-world document processing tasks and helps developers prioritize improvements in long-context Korean language comprehension.