Flair-abbr-pubmed-unfiltered

This model is a fine-tuned version of HunFlair pubmed models on the PLODv2 unfiltered dataset. It is released with our LREC-COLING 2024 publication Using character-level models for efficient abbreviation and long-form detection. It achieves the following results on the test set:

Results on abbreviations:

• Precision: 0.8502
• Recall: 0.9155
• F1: 0.8816

Results on long forms:

• Precision: 0.8270
• Recall: 0.8864
• F1: 0.8557