Natural Language Processing Section, Department of Computer Science, University of Copenhagen

CopeNLU is a Natural Language Processing research group led by Isabelle Augenstein with a focus on researching methods for tasks that require a deep understanding of language, as opposed to shallow processing. We are affiliated with the Natural Language Processing Section at the Department of Computer Science, University of Copenhagen. We are interested in core methodology research on, among others, learning with limited training data and explainable AI; as well as applications thereof to tasks such as fact checking, gender bias detection and question answering. Our group is partly funded by an ERC Starting Grant on Explainable and Robust Automatic Fact Checking, as well as a Sapere Aude Research Leader fellowship on `Learning to Explain Attitudes on Social Media’.

Interests

Natural Language Understanding
Learning with Limited Labelled Data
Explainable AI
Fact Checking
Question Answering
Gender Bias Detection

News

More Talks

4 Papers Accepted to ACL 2023

22 May 2023

61st Annual Meeting of the Association for Computational Linguistics (ACL 2023)

4 papers by CopeNLU authors are accepted to appear at ACL 2023, on faithfulness of explanations, measuring intersectional biases, event extraction and few-shot stance detection.

PhD position available in context of Carlsberg-funded project on bias detection in job ads

27 Jan 2023

Carlsberg PhD Fellowship Call

1 PhD fellowship on fair and accountable natural language processing is available in CopeNLU. The positions are funded by a Carlsberg Semper Ardens project on detecing biases in job ads. The application deadline is 24 May 2023.

Project Project

Positions available in context of ERC Starting Grant project ExplainYourself

22 Nov 2022

ERC StG PhD and Postdoc Fellowship Call

PhD and postdoctoral fellowships on explainable fact checking are available in CopeNLU. The positions are funded by the ERC Starting Grant project ExplainYourself. The application deadline for the postdoctoral position is 24 May 2023.

Project Project

Isabelle Augenstein becomes Denmark's youngest female full professor

17 Oct 2022

Denmark’s youngest female professor appointed at the University of Copenhagen

Isabelle Augenstein was promoted to full professor, making her the youngest ever female full professor in Denmark.

2 Papers Accepted to Coling 2022

12 Oct 2022

The 29th International Conference on Computational Linguistics (Coling 2022)

2 papers by CopeNLU authors are accepted to appear at Coling 2022, on probing of question answering models.

2 Papers Accepted to EMNLP 2022

1 Oct 2022

The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022)

2 papers by CopeNLU authors are accepted to appear at EMNLP 2022, on scientific document understanding.

3 Papers Accepted to NAACL 2022

10 Jul 2022

2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022)

3 papers by CopeNLU authors are accepted to appear at NAACL 2022, on hatespeech detection, misinformation detection and multilingual probing.

2 Papers Accepted to AAAI 2022

1 Dec 2021

Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI 2022)

2 papers by CopeNLU authors are accepted to appear at AAAI 2022, on explanation generation, as well as on cross-lingual stance detection.

EXPANSE Project Kick-Off

1 Sep 2021

DFF Sapere Aude project EXPANSE

On 1 September 2021, the DFF Sapere Aude project EXPANSE on ‘Learning to Explain Attitudes on Social Media’ is kicking off, and four new members are joining our group.

Project Project

3 Papers Accepted to EMNLP 2021

26 Aug 2021

2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021)

3 papers by CopeNLU authors are accepted to appear at EMNLP 2021, on stance detection, exaggeration detection, and on counterfactually augmented data.

People

Isabelle Augenstein

Full Professor

Isabelle’s main research interests are natural language understanding, explainability and learning with limited training data.

Pepa Atanasova

Postdoc

Pepa’s research interests are multilingual fact checking and question answering.

Nils Rethmeier

PhD Student

Nils researches low-resource learning and unsupervised learning as well as explainability.

Karolina Stanczak

PhD Student

Karolina’s research interests include gender-biased language detection and statistical methods.

Erik Arakelyan

PhD Student

Erik’s main research interests are question answering and explainability.

Nadav Borenstein

PhD Student

Nadav’s research interests include improving the trustworthiness and usefulness of deep models in the NLP domain.

Arnav Arora

PhD Student

Arnav’s research interests include equitable ML, mitigating online harms, and the intersection of NLP and Computational Social Science.

Marta Marchiori Manerba

PhD Intern

Marta’s research interests are explainability and fairness, with a focus on abusive language detection. She is a second-year PhD student at the University of Pisa and is visiting CopeNLU in Spring/Summer 2023.

Kevin Roitero

Visiting Researcher

Kevin Roitero is an Assistant Professor (RTD/a) at the University of Udine, Italy. His research interests include Artificial Intelligence, Deep Learning, Crowdsourcing, and Information Retrieval. He is visiting CopeNLU in Spring/Summer 2023.

Ryan Cotterell

Lecturer

This is Ryan. He’s a lecturer at the University of Cambridge and a frequent collaborator of the CopeNLU group.

Oscar Kjell

Postdoc

Oscar researches automating the assessment of psychological constructs using representation learning. He is an international postdoc affiliated with Lund University as well the University of Copenhagen and Stony Brook University.

Sara Vera Marjanovic

PhD Student

Sara’s research interests include explainable IR and NLP models, identifying biases in large text datasets, as well as working with social media data. She is a member of the DIKU ML section and IR group and co-advised by Isabelle.

Amelie Wührl

PhD Intern

Amelie is a PhD Student at the University of Stuttgart, interested in fact-checking, specifically for medical claims and content that people chose to share on social media. She was visiting CopeNLU in Spring 2023 to work on scientific fact checking.

Lucie-Aimée Kaffee

Postdoc

Lucie’s research interests include supporting lower-resourced language communities (including Wikipedia and Wikidata) with NLP, and multilingual knowledge graphs. She was a postdoctoral researcher in the CopeNLU group from 2021-2023 and is now a postdoctoral researcher at Hasso-Plattner-Institut.

Andreas Nugaard Holm

PhD Student

Andreas’ main research areas are representation learning and domain adaptation, with a focus on scientific texts.

Dustin Wright

PhD Student

Dustin’s research interests include fact checking and knowledge base population, with a focus on scientific texts. He was PhD in CopeNLU from 2019 to 2023, and is now a postdoctoral researcher in the DIKU ML section.

Thea Rolskov Sloth

Intern

Thea is a MSc Student in Cognitive Science at Aarhus University, visiting CopeNLU in Autumn 2022 to work on gender bias detection.

Miryam de Lhoneux

Postdoc

Miryam’s main research interests are syntactic parsing, multilingual NLP and interpretability. She is an assistant profesor at KU Leuven, and was previously an international postdoc affiliated with KU Leuven as well as the University of Copenhagen.

Nodens Koren

PhD Student

Nodens is interested in interpretability and bio-inspired models.

Klim Zaporojets

PhD Intern

Klim is a PhD Student at Ghent University. He was visiting CopeNLU in Spring 2022 to work on entity linking.

Sagnik Ray Choudhury

Postdoc

Sagnik is a postdoctoral researcher at the University of Michigan. He was a postdoctoral researcher in CopeNLU from 2020 to 2022, working on question answering and interpretability of neural black-box models.

Shailza Jolly

PhD Intern

Shailza is a PhD Student at the Technical University of Kaiserslautern, and a research assistant at DFKI. She was visiting CopeNLU in Winter/Spring 2021 to work on interpretability.

Yova Kementchedjhieva

PhD Student

Yova was a PhD student in the CoAStaL NLP group researching low-resource and cross-lingual learning, co-advised by Isabelle. She is now a postdoc in the same group.

Ana Valeria González

PhD Student

Ana is a senior data scientist at Halfspace, Copenhagen. Prior to this, she was a PhD student in the CoAStaL NLP group from 2017-2021 co-advised by Isabelle, working on question answering.

Andrea Lekkas

PhD Student

Andrea’s main research interests are multilingual learning and language modelling. He was a PhD student at the University of Southern Denmark and co-advised by Isabelle.

Johannes Bjerva

Postdoc

Johannes is an associate professor at Aalborg University Copenhagen. He was a postdoc in CopeNLU from 2017 to 2020, researching multi-lingual and multi-task learning.

Liesbeth Allein

PhD Intern

Liesbeth is a PhD Student at KU Leuven, and was visiting CopeNLU in Spring 2020 to work on fact checking.

Wei Zhao

PhD Intern

Wei Zhao is a postdoctoral researcher at Heidelberg University, and was visiting CopeNLU in Winter 2019 to work on low-resource natural language generation.

Mareike Hartmann

PhD Student

Mareike was a member of the CoAStaL NLP group, researching disinformation detection and co-advised by Isabelle, and is now a researcher at DFKI.

Farhad Nooralahzadeh

PhD Intern

Farhad is a currently postdoc at the University of Zurich, and was visiting CopeNLU in Spring 2019 to work on domain adaptation for information extraction.

Zhong Xuan

Research Intern

Zhong Xuan is a data scientist at Grab, Singapore, and was visiting CopeNLU in Summer 2019 to work on relation extraction and knowledge base population.

Luna De Bruyne

PhD Intern

Luna is a PhD Student at Ghent University, and was visiting CopeNLU in Spring 2019 to work on emotion detection.

Giannis Bekoulis

PhD Intern

Giannis is a principal data scientist at Clarivate, and was visiting CopeNLU as a PhD Intern in Spring 2019 to work on joint information extraction.

Sune Debel

Research Assistant

Now a machine learning engineer at Omhu.

Featured Publications

Fact Checking with Insufficient Evidence

Automating the fact checking (FC) process relies on information obtained from external sources. In this work, we posit that it is crucial for FC models to make veracity predictions only when there is sufficient evidence and otherwise indicate when it is not enough. To this end, we are the first to study what information FC models consider sufficient by introducing a novel task and advancing it with three main contributions. First, we conduct an in-depth empirical analysis of the task with a new fluency-preserving method for omitting information from the evidence at the constituent and sentence level. We identify when models consider the remaining evidence (in)sufficient for FC, based on three trained models with different Transformer architectures and three FC datasets. Second, we ask annotators whether the omitted evidence was important for FC, resulting in a novel diagnostic dataset, SufficientFacts, for FC with omitted evidence. We find that models are least successful in detecting missing evidence when adverbial modifiers are omitted (21% accuracy), whereas it is easiest for omitted date modifiers (63% accuracy). Finally, we propose a novel data augmentation strategy for contrastive self-learning of missing evidence by employing the proposed omission method combined with tri-training. It improves performance for Evidence Sufficiency Prediction by up to 17.8 F1 score, which in turn improves FC performance by up to 2.6 F1 score.

Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

In TACL, 2022

PDF Project Project

Generating Fact Checking Explanations

This paper provides the first study of how fact checking explanations can be generated automatically based on available claim context, and how this task can be modeled jointly with veracity prediction.

Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

In ACL, 2020

Preprint Project Project Video

Recent Publications

More Publications

Faithfulness Tests for Natural Language Explanations

Explanations of neural models aim to reveal a model’s decision-making process for its predictions. However, recent work shows …

Pepa Atanasova, Oana-Maria Camburu, Christina Lioma, Thomas Lukasiewicz, Jakob Grue Simonsen, Isabelle Augenstein

PDF Project

Measuring Intersectional Biases in Historical Documents

Data-driven analyses of biases in historical texts can help illuminate the origin and development of biases prevailing in modern …

Nadav Borenstein, Karolina Stańczak, Thea Rolskov, Natacha Klein Käfer, Natália da Silva Perez, Isabelle Augenstein

PDF Project

Multilingual Event Extraction from Historical Newspaper Adverts

NLP methods can aid historians in analyzing textual materials in greater volumes than manually feasible. Developing such methods poses …

Nadav Borenstein, Natália da Silva Perez, Isabelle Augenstein

PDF Project

Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection

The task of Stance Detection is concerned with identifying the attitudes expressed by an author towards a target of interest. This task …

Erik Arakelyan, Arnav Arora, Isabelle Augenstein

PDF Project Project

Measuring Gender Bias in West Slavic Language Models

Pre-trained language models have been known to perpetuate biases from the underlying datasets to downstream tasks. However, these …

Sandra Martinková, Karolina Stańczak, Isabelle Augenstein

PDF Project Project

Probing Pre-Trained Language Models for Cross-Cultural Differences in Values

Language embeds information about social, cultural, and political values people hold. Prior work has explored social and potentially …

Arnav Arora, Lucie-Aimée Kaffee, Isabelle Augenstein

PDF Project Project

Thorny Roses: Investigating the Dual Use Dilemma in Natural Language Processing

Dual use, the intentional, harmful reuse of technology and scientific artefacts, is a problem yet to be well-defined within the context …

Lucie-Aimée Kaffee, Arnav Arora, Isabelle Augenstein

PDF

Adapting Neural Link Predictors for Complex Query Answering

Answering complex queries on incomplete knowledge graphs is a challenging task where a model needs to answer complex logical queries in …

Erik Arakelyan, Pasquale Minervini, Isabelle Augenstein

PDF Project Project Project

Multi-View Knowledge Distillation from Crowd Annotations for Out-of-Domain Generalization

Selecting an effective training signal for tasks in natural language processing is difficult: expert annotations are expensive, and …

Dustin Wright, Isabelle Augenstein

PDF Project

A Latent-Variable Model for Intrinsic Probing

The success of pre-trained contextualized representations has prompted researchers to analyze them for the presence of linguistic …

Karolina Stańczak, Lucas Torroba Hennigen, Adina Williams, Ryan Cotterell, Isabelle Augenstein

PDF Project Project

Interests

News

People

Full Professor

Postdoc

PhD Student

PhD Student

PhD Student

PhD Student

PhD Student

PhD Intern

Visiting Researcher

Lecturer

Postdoc

PhD Student

PhD Intern

Postdoc

PhD Student

PhD Student

Intern

Postdoc

PhD Student

PhD Intern

Postdoc

PhD Intern

PhD Student

PhD Student

PhD Student

Postdoc

PhD Intern

PhD Intern

PhD Student

PhD Intern

Research Intern

PhD Intern

PhD Intern

Research Assistant

Featured Publications

Recent Publications

Recent Posts

Projects

Contact