EQUES

/

jpharma-bert-large

like 1

Follow

EQUES Inc. 36

This PR improves the model card for EQUES/jpharma-bert-large by:

• Adding pipeline_tag: fill-mask to the metadata, which ensures the model is properly categorized and enables the correct interactive widget on the Hugging Face Hub (e.g., at https://huggingface.co/models?pipeline_tag=fill-mask). This aligns with the provided sample usage (pipeline("fill-mask", ...)).
• Specifying language: [ja, en] as the model was continually pre-trained on both Japanese and English pharmaceutical/biomedical tokens, as stated in the paper abstract.
• Including relevant tags such as bert, japanese, pharmaceutical, and biomedical for better discoverability and categorization.
• Adding a prominent link to the paper A Japanese Language Model and Three New Evaluation Benchmarks for Pharmaceutical NLP at the top of the model card.
• Correcting the typo "Examoke Usage" to "Example Usage".