Allan Hanbury

AustroTox: A Dataset for Target-Based Austrian German Offensive Language Detection

2024 · arXiv (Cornell University)

Model interpretability in toxicity detection greatly profits from token-level annotations. However, currently such annotations are only available in English. We introduce a dataset annotated for offensive language detection sourced from a news forum, notable for …

Relation Extraction or Pattern Matching? Unravelling the Generalisation Limits of Language Models for Biographical RE

2025 · arXiv (Cornell University)

Analysing the generalisation capabilities of relation extraction (RE) models is crucial for assessing whether they learn robust relational patterns or rely on spurious correlations. Our cross-dataset experiments find that RE models struggle with unseen data, …