albert-kor-base is an ALBERT model pretrained on a 70GB Korean text corpus using 42,000 lower-cased subword tokens. It supports fill-mask tasks and can be used with the Hugging Face Transformers library for various Korean downstream NLP applications such as text classification, named entity recognition, and question answering.
Albert Kor Base sits in PulseGate's Foundation models & chat category. It focuses on providing a high-quality pretrained ALBERT model specifically for Korean natural language processing. It is built as an open-source project for Korean NLP developers and researchers. Albert Kor Base is open source under the Apache-2.0 license. It runs on the web and API.
kykim builds and maintains Albert Kor Base, and the product first shipped in 2020. Development happens publicly on GitHub with 394 stars. Key capabilities include fill-mask, korean language support, and ALBERT architecture.
Latest indexed changes and source events
kykim/albert-kor-base verified by the PulseGate indexer
