PLASSLeonhard Hennig, Christoph AltFeb 23, 2021Go to Project SiteInformation Extraction Low-Resource LearningLeonhard HennigSenior ResearcherChristoph AltRelatedData4TransparencyText2TechBIFOLDCora4NLPSIM3SPublicationsMultiTACRED: A Multilingual Version of the TAC Relation Extraction DatasetRelation extraction (RE) is a fundamental task in information extraction, whose extension to multilingual settings has been hindered by …Leonhard Hennig, Philippe Thomas, Sebastian MöllerPDF Cite Code Dataset Project Project Project DOIA Comparative Study of Pre-trained Encoders for Low-Resource Named Entity RecognitionPre-trained language models (PLM) are effective components of few-shot named entity recognition (NER) approaches when augmented with …Yuxuan Chen, Jonas Mikkelsen, Arne Binder, Christoph Alt, Leonhard HennigPDF Cite Code Project ProjectWhy only Micro-$F_1$? Class Weighting of Measures for Relation ClassificationRelation classification models are conventionally evaluated using only a single measure, e.g., micro-F1, macro-F1 or AUC. In this work, …David Harbecke, Yuxuan Chen, Leonhard Hennig, Christoph AltCite Project ProjectDetecting Covariate Drift with ExplanationsDetecting when there is a domain drift between training and inference data is important for any model evaluated on data collected in …Steffen Castle, Robert Schwarzenberg, Mohsen PourvaliCite Code Project DOIDefx at SemEval-2020 Task 6: Joint Extraction of Concepts and Relations for Definition ExtractionWe describe our submissions to the DeftEval shared task (SemEval-2020 Task 6)Marc Hübner, Christoph Alt, Robert Schwarzenberg, Leonhard HennigPDF Cite Code ProjectBootstrapping Named Entity Recognition in E-Commerce with Positive Unlabeled LearningIn this work, we introduce a bootstrapped, iterative NER model that integrates a PU learning algorithm for recognizing named entities …Hanchu Zhang, Leonhard Hennig, Christoph Alt, Changjian Hu, Yao Meng, Chao WangPDF Cite Project DOIProbing Linguistic Features of Sentence-Level Representations in Neural Relation ExtractionDespite the recent progress, little is known about the features captured by state-of-the-art neural relation extraction (RE) models. …Christoph Alt, Aleksandra Gabryszak, Leonhard HennigPDF Cite Code Project Project Project DOITACRED Revisited: A Thorough Evaluation of the TACRED Relation Extraction TaskTACRED is one of the largest, most widely used crowdsourced datasets in Relation Extraction (RE). But, even with recent advances in …Christoph Alt, Aleksandra Gabryszak, Leonhard HennigPDF Cite Code Project Project Project DOI
