Saved in:
Bibliographic Details
Main Authors: Jalocha, Maciej, Schmidt, Johan Hausted, Michelseen, William
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.13870
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918133356298240
author Jalocha, Maciej
Schmidt, Johan Hausted
Michelseen, William
author_facet Jalocha, Maciej
Schmidt, Johan Hausted
Michelseen, William
contents The field of cybersecurity NER lacks standardized labels, making it challenging to combine datasets. We investigate label unification across four cybersecurity datasets to increase data resource usability. We perform a coarse-grained label unification and conduct pairwise cross-dataset evaluations using BiLSTM models. Qualitative analysis of predictions reveals errors, limitations, and dataset differences. To address unification limitations, we propose alternative architectures including a multihead model and a graph-based transfer model. Results show that models trained on unified datasets generalize poorly across datasets. The multihead model with weight sharing provides only marginal improvements over unified training, while our graph-based transfer model built on BERT-base-NER shows no significant performance gains compared BERT-base-NER.
format Preprint
id arxiv_https___arxiv_org_abs_2507_13870
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Label Unification for Cross-Dataset Generalization in Cybersecurity NER
Jalocha, Maciej
Schmidt, Johan Hausted
Michelseen, William
Computation and Language
The field of cybersecurity NER lacks standardized labels, making it challenging to combine datasets. We investigate label unification across four cybersecurity datasets to increase data resource usability. We perform a coarse-grained label unification and conduct pairwise cross-dataset evaluations using BiLSTM models. Qualitative analysis of predictions reveals errors, limitations, and dataset differences. To address unification limitations, we propose alternative architectures including a multihead model and a graph-based transfer model. Results show that models trained on unified datasets generalize poorly across datasets. The multihead model with weight sharing provides only marginal improvements over unified training, while our graph-based transfer model built on BERT-base-NER shows no significant performance gains compared BERT-base-NER.
title Label Unification for Cross-Dataset Generalization in Cybersecurity NER
topic Computation and Language
url https://arxiv.org/abs/2507.13870