Dictionary-MKG is a next-generation multilingual dataset designed to bridge the gap between static dictionaries and dynamic language learning. Generated using state-of-the-art LLMs (currently gemini-3-flash-preview), this project aims to provide structured, high-quality learning resources for language pairs that are historically under-served (e.g., learning Korean through Spanish).
You can find an… See the full description on the dataset page:
https://huggingface.co/datasets/kenantang/Dictionary-MKG.
