3 papers by CopeNLU authors are accepted to appear at EMNLP 2021. The topics of these include stance detection, exaggeration detection, and counterfactually augmented data.
Cross-Domain Label-Adaptive Stance Detection. Momchil Hardalov, Arnav Arora, Preslav Nakov, Isabelle Augenstein. In Proceedings of EMNLP.
How Does Counterfactually Augmented Data Impact Models for Social Computing Constructs?. Indira Sen, Mattia Samory, Fabian Flöck, Claudia Wagner, Isabelle Augenstein. In Proceedings of EMNLP.
Semi-Supervised Exaggeration Detection of Health Science Press Releases.
A paper by CopeNLU author is accepted to appear at IJCAI 2021. The paper studies how to perform complex claim verification on naturally occurring political claims with multiple hops over evidence chunks.
Multi-Hop Fact Checking of Political Claims. Wojciech Ostrowski, Arnav Arora, Pepa Atanasova, Isabelle Augenstein.
2 papers by CopeNLU authors are accepted to appear at ACL 2021. One paper is on interpretability, examining how sparsity affects our ability to use attention as an explainability tool; whereas the other one is on scientific document understanding, introducing a new dataset for the task of cite-worthiness detection in scientific articles.
Is Sparse Attention more Interpretable? Clara Meister, Stefan Lazov, Isabelle Augenstein, Ryan Cotterell.
CiteWorth: Cite-Worthiness Detection for Improved Scientific Document Understanding.
A paper by CopeNLU author is accepted to appear at EACL 2021. The paper aims to bridge the gap between high- and low-resource languages by investigating to what degree cross-lingual models share structural information about languages.
Does Typological Blinding Impede Cross-Lingual Sharing?. Johannes Bjerva, Isabelle Augenstein.
Two PhD fellowships and two postdoc positions on explainable stance detection are available in CopeNLU. The PhD fellowships and one of the postdoc positions are offered in the context of a DFF Sapere Aude research leader fellowship on `Learning to Explain Attitudes on Social Media (EXPANSE)´. Sapere Aude is a program by the Independent Research Fund Denmark (DFF) to support the most talented younger researchers in Denmark with funding for blue-skies research to build up or expand their research groups.
7 CopeNLU papers are accepted to appear at EMNLP 2020. The topics of these include fact checking, explainability, domain adaptation, transfer learning, QA and improving peer review.
A Diagnostic Study of Explainability Techniques for Text Classification. Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein. In Proceedings of EMNLP.
Generating Label Cohesive and Well-Formed Adversarial Claims. Pepa Atanasova*, Dustin Wright*, Isabelle Augenstein. In Proceedings of EMNLP.
Transformer Based Multi-Source Domain Adaptation.
2 papers by CopeNLU authors are accepted to appear at ACL 2020. One paper is on explainable fact checking, providing the first study of how fact checking explanations can be generated automatically based on claim content, and how this task can be modelled jointly with veracity prediction; whereas the other one is on script conversion, proposing a novel Chinese character conversion model that can disambiguate between mappings and convert between Chinese scripts.
4 papers by CopeNLU authors are to be presented at EMNLP 2019 and co-located events, on fact checking and disinformation, as well as on multi-task and multi-lingual learning.
MultiFC: A Real-World Multi-Domain Dataset for Evidence-Based Fact Checking of Claims. Isabelle Augenstein, Christina Lioma, Dongsheng Wang, Lucas Chaves Lima, Casper Hansen, Christian Hansen, Jakob Grue Simonsen. In Proceedings of EMNLP-IJCNLP 2019.
Mapping (Dis-)Information Flow about the MH17 Plane Crash. Mareike Hartmann, Yevgeniy Golovchenko, Isabelle Augenstein.
2 papers by CopeNLU authors are accepted to appear at ACL 2019. One paper is on uncovering probabilistic implications in typological knowledge bases, following up from our NAACL 2019 paper on generative linguistic typology; whereas the other one is on unsupervised discovery of gendered language, utilising the multi-view autoencoder introduced in our NAACL 219 paper.
Uncovering Probabilistic Implications in Typological Knowledge Bases. Johannes Bjerva, Yova Kementchedjhieva, Ryan Cotterell, Isabelle Augenstein.
3 papers by CopeNLU authors are accepted to appear at NAACL 2019. Topics span from population of typological knowledge bases and weak supervision from disparate lexica to frame detection in online fora.
A Probabilistic Generative Model of Linguistic Typology. Johannes Bjerva, Yova Kementchedjhieva, Ryan Cotterell, Isabelle Augenstein.
Combining Disparate Sentiment Lexica with a Multi-View Variational Autoencoder. Alexander Hoyle, Lawrence Wolf-Sonkin, Hanna Wallach, Ryan Cotterell, Isabelle Augenstein.
Issue Framing in Online Discussion Fora.