SubmissionNumber#=%=#44
FinalPaperTitle#=%=#Effective Multi-Task Learning for Biomedical Named Entity Recognition
ShortPaperTitle#=%=#
NumberOfPages#=%=#15
CopyrightSigned#=%=#João Ruano
JobTitle#==#
Organization#==#
Abstract#==#Biomedical Named Entity Recognition presents significant challenges due to the complexity of biomedical terminology and inconsistencies in annotation across datasets. This paper introduces SRU-NER (Slot-based Recurrent Unit NER), a novel approach designed to handle nested named entities while integrating multiple datasets through an effective multi-task learning strategy. SRU-NER mitigates annotation gaps by dynamically adjusting loss computation to avoid penalizing predictions of entity types absent in a given dataset. Through extensive experiments, including a cross-corpus evaluation and human assessment of the model's predictions, SRU-NER achieves competitive performance in biomedical and general-domain NER tasks, while improving cross-domain generalization.
Author{1}{Firstname}#=%=#João
Author{1}{Lastname}#=%=#Ruano
Author{1}{Username}#=%=#joaoruano
Author{1}{Email}#=%=#joao.ruano.work@gmail.com
Author{1}{Affiliation}#=%=#Priberam
Author{2}{Firstname}#=%=#Gonçalo M.
Author{2}{Lastname}#=%=#Correia
Author{2}{Username}#=%=#goncalomcorreia
Author{2}{Email}#=%=#goncalommac@gmail.com
Author{2}{Affiliation}#=%=#Priberam
Author{3}{Firstname}#=%=#Leonor Maria Machado
Author{3}{Lastname}#=%=#Barreiros
Author{3}{Username}#=%=#leonormbarreiros
Author{3}{Email}#=%=#leonormbarreiros@gmail.com
Author{3}{Affiliation}#=%=#Priberam
Author{4}{Firstname}#=%=#Afonso
Author{4}{Lastname}#=%=#Mendes
Author{4}{Username}#=%=#afonsoamendes
Author{4}{Email}#=%=#amm@priberam.pt
Author{4}{Affiliation}#=%=#Priberam Informática, SA.

==========
èéáğö