Search Results for author: Isabelle Augenstein

Found 78 papers, 29 papers with code

Can Edge Probing Tasks Reveal Linguistic Knowledge in QA Models?

no code implementations15 Sep 2021 Sagnik Ray Choudhury, Nikita Bhutani, Isabelle Augenstein

There have been many efforts to try to understand what grammatical knowledge (e. g., ability to understand the part of speech of a token) is encoded in large pre-trained language models (LM).

Question Answering

Few-Shot Cross-Lingual Stance Detection with Sentiment-Based Pre-Training

no code implementations13 Sep 2021 Momchil Hardalov, Arnav Arora, Preslav Nakov, Isabelle Augenstein

Most research in stance detection, however, has been limited to working with a single language and on a few limited targets, with little work on cross-lingual stance detection.

Stance Detection

Diagnostics-Guided Explanation Generation

no code implementations8 Sep 2021 Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

When such annotations are not available, explanations are often selected as those portions of the input that maximise a downstream task's performance, which corresponds to optimising an explanation's Faithfulness to a given model.

Semi-Supervised Exaggeration Detection of Health Science Press Releases

1 code implementation30 Aug 2021 Dustin Wright, Isabelle Augenstein

Given this, we present a formalization of and study into the problem of exaggeration detection in science communication.

Few-Shot Learning

Towards Explainable Fact Checking

no code implementations23 Aug 2021 Isabelle Augenstein

This development has spurred research in the area of automatic fact checking, from approaches to detect check-worthy claims and determining the stance of tweets towards claims, to methods to determine the veracity of claims given evidence documents.

Decision Making Fact Checking +2

QA Dataset Explosion: A Taxonomy of NLP Resources for Question Answering and Reading Comprehension

no code implementations27 Jul 2021 Anna Rogers, Matt Gardner, Isabelle Augenstein

Question answering and reading comprehension have been particularly prolific in this regard, with over 80 new datasets appearing in the past two years.

Question Answering Reading Comprehension

Is Sparse Attention more Interpretable?

no code implementations ACL 2021 Clara Meister, Stefan Lazov, Isabelle Augenstein, Ryan Cotterell

Sparse attention has been claimed to increase model interpretability under the assumption that it highlights influential inputs.

Text Classification

Determining the Credibility of Science Communication

no code implementations30 May 2021 Isabelle Augenstein

Most work on scholarly document processing assumes that the information processed is trustworthy and factually correct.

Quantifying Gender Bias Towards Politicians in Cross-Lingual Language Models

no code implementations15 Apr 2021 Karolina Stańczak, Sagnik Ray Choudhury, Tiago Pimentel, Ryan Cotterell, Isabelle Augenstein

While the prevalence of large pre-trained language models has led to significant improvements in the performance of NLP systems, recent research has demonstrated that these models inherit societal biases extant in natural language.

Language Modelling

Cross-Domain Label-Adaptive Stance Detection

no code implementations15 Apr 2021 Momchil Hardalov, Arnav Arora, Preslav Nakov, Isabelle Augenstein

In this paper, we perform an in-depth analysis of 16 stance detection datasets, and we explore the possibility for cross-domain learning from them.

Domain Adaptation Stance Detection

A Neighbourhood Framework for Resource-Lean Content Flagging

no code implementations31 Mar 2021 Sheikh Muhammad Sarwar, Dimitrina Zlatkova, Momchil Hardalov, Yoan Dinkov, Isabelle Augenstein, Preslav Nakov

We propose a novel interpretable framework for cross-lingual content flagging, which significantly outperforms prior work both in terms of predictive performance and average inference time.

Abusive Language

University of Copenhagen Participation in TREC Health Misinformation Track 2020

no code implementations3 Mar 2021 Lucas Chaves Lima, Dustin Brandon Wright, Isabelle Augenstein, Maria Maistro

Our approach consists of 3 steps: (1) we create an initial run with BM25 and RM3; (2) we estimate credibility and misinformation scores for the documents in the initial run; (3) we merge the relevance, credibility and misinformation scores to re-rank documents in the initial run.

Language Modelling Misinformation +1

Detecting Abusive Language on Online Platforms: A Critical Analysis

no code implementations27 Feb 2021 Preslav Nakov, Vibha Nayak, Kyle Dent, Ameya Bhatawdekar, Sheikh Muhammad Sarwar, Momchil Hardalov, Yoan Dinkov, Dimitrina Zlatkova, Guillaume Bouchard, Isabelle Augenstein

Abusive language on online platforms is a major societal problem, often leading to important societal problems such as the marginalisation of underrepresented minorities.

Abusive Language

A Survey on Stance Detection for Mis- and Disinformation Identification

no code implementations27 Feb 2021 Momchil Hardalov, Arnav Arora, Preslav Nakov, Isabelle Augenstein

Detecting attitudes expressed in texts, also known as stance detection, has become an important task for the detection of false information online, be it misinformation (unintentionally false) or disinformation (intentionally false, spread deliberately with malicious intent).

Fact Checking Misinformation +3

A Primer on Contrastive Pretraining in Language Processing: Methods, Lessons Learned and Perspectives

no code implementations25 Feb 2021 Nils Rethmeier, Isabelle Augenstein

Contrastive self-supervised training objectives enabled recent successes in image representation pretraining by learning to contrast input-input pairs of augmented images as either similar or dissimilar.

Contrastive Learning Language Modelling +2

Does Typological Blinding Impede Cross-Lingual Sharing?

no code implementations EACL 2021 Johannes Bjerva, Isabelle Augenstein

Our hypothesis is that a model trained in a cross-lingual setting will pick up on typological cues from the input data, thus overshadowing the utility of explicitly using such features.

Disembodied Machine Learning: On the Illusion of Objectivity in NLP

no code implementations28 Jan 2021 Zeerak Waseem, Smarika Lulz, Joachim Bingel, Isabelle Augenstein

In this paper, we contextualise this discourse of bias in the ML community against the subjective choices in the development process.

Longitudinal Citation Prediction using Temporal Graph Neural Networks

no code implementations10 Dec 2020 Andreas Nugaard Holm, Barbara Plank, Dustin Wright, Isabelle Augenstein

Citation count prediction is the task of predicting the number of citations a paper has gained after a period of time.

Citation Prediction

Multi-Sense Language Modelling

no code implementations10 Dec 2020 Andrea Lekkas, Peter Schneider-Kamp, Isabelle Augenstein

The effectiveness of a language model is influenced by its token representations, which must encode contextual information and handle the same word form having a plurality of meanings (polysemy).

Graph Attention Language Modelling +1

What Can We Do to Improve Peer Review in NLP?

no code implementations Findings of the Association for Computational Linguistics 2020 Anna Rogers, Isabelle Augenstein

Peer review is our best tool for judging the quality of conference submissions, but it is becoming increasingly spurious.

Unsupervised Evaluation for Question Answering with Transformers

no code implementations7 Oct 2020 Lukas Muttenthaler, Isabelle Augenstein, Johannes Bjerva

We observe a consistent pattern in the answer representations, which we show can be used to automatically evaluate whether or not a predicted answer span is correct.

Question Answering

Data-Efficient Pretraining via Contrastive Self-Supervision

no code implementations2 Oct 2020 Nils Rethmeier, Isabelle Augenstein

For natural language processing `text-to-text' tasks, the prevailing approaches heavily rely on pretraining large self-supervised models on increasingly larger `task-external' data.

Fairness Few-Shot Learning +3

A Diagnostic Study of Explainability Techniques for Text Classification

1 code implementation EMNLP 2020 Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

Recent developments in machine learning have introduced models that approach human performance at the cost of increased architectural complexity.

Classification General Classification +1

Generating Label Cohesive and Well-Formed Adversarial Claims

1 code implementation EMNLP 2020 Pepa Atanasova, Dustin Wright, Isabelle Augenstein

However, for inference tasks such as fact checking, these triggers often inadvertently invert the meaning of instances they are inserted in.

Fact Checking Language Modelling +1

Transformer Based Multi-Source Domain Adaptation

1 code implementation EMNLP 2020 Dustin Wright, Isabelle Augenstein

Here, we investigate the problem of unsupervised multi-source domain adaptation, where a model is trained on labelled data from multiple source domains and must make predictions on a domain for which no labelled data has been seen.

Domain Adaptation

Multi-Hop Fact Checking of Political Claims

1 code implementation10 Sep 2020 Wojciech Ostrowski, Arnav Arora, Pepa Atanasova, Isabelle Augenstein

We: 1) construct a small annotated dataset, PolitiHop, of evidence sentences for claim verification; 2) compare it to existing multi-hop datasets; and 3) study how to transfer knowledge from more extensive in- and out-of-domain resources to PolitiHop.

Fact Checking Transfer Learning

2kenize: Tying Subword Sequences for Chinese Script Conversion

1 code implementation ACL 2020 Pranav A, Isabelle Augenstein

Simplified Chinese to Traditional Chinese character conversion is a common preprocessing step in Chinese NLP.

General Classification Topic Classification

SubjQA: A Dataset for Subjectivity and Review Comprehension

1 code implementation EMNLP 2020 Johannes Bjerva, Nikita Bhutani, Behzad Golshan, Wang-Chiew Tan, Isabelle Augenstein

We find that subjectivity is also an important feature in the case of QA, albeit with more intricate interactions between subjectivity and QA performance.

Question Answering Sentiment Analysis +1

Generating Fact Checking Explanations

no code implementations ACL 2020 Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

Most existing work on automated fact checking is concerned with predicting the veracity of claims based on metadata, social network spread, language used in claims, and, more recently, evidence supporting or denying claims.

Fact Checking

Zero-Shot Cross-Lingual Transfer with Meta Learning

1 code implementation EMNLP 2020 Farhad Nooralahzadeh, Giannis Bekoulis, Johannes Bjerva, Isabelle Augenstein

We show that this challenging setup can be approached using meta-learning, where, in addition to training a source language model, another model learns to select which training instances are the most beneficial to the first.

Language Modelling Meta-Learning +4

TX-Ray: Quantifying and Explaining Model-Knowledge Transfer in (Un-)Supervised NLP

2 code implementations2 Dec 2019 Nils Rethmeier, Vageesh Kumar Saxena, Isabelle Augenstein

While state-of-the-art NLP explainability (XAI) methods focus on explaining per-sample decisions in supervised end or probing tasks, this is insufficient to explain and quantify model knowledge transfer during (un-)supervised training.

Model Compression Transfer Learning

Joint Emotion Label Space Modelling for Affect Lexica

no code implementations20 Nov 2019 Luna De Bruyne, Pepa Atanasova, Isabelle Augenstein

Emotion lexica are commonly used resources to combat data poverty in automatic emotion detection.

Emotion Recognition

Mapping (Dis-)Information Flow about the MH17 Plane Crash

1 code implementation WS 2019 Mareike Hartmann, Yevgeniy Golovchenko, Isabelle Augenstein

In this work, we examine to what extent text classifiers can be used to label data for subsequent content analysis, in particular we focus on predicting pro-Russian and pro-Ukrainian Twitter content related to the MH17 plane crash.

Retrieval-based Goal-Oriented Dialogue Generation

no code implementations30 Sep 2019 Ana Valeria Gonzalez, Isabelle Augenstein, Anders Søgaard

Most research on dialogue has focused either on dialogue generation for openended chit chat or on state tracking for goal-directed dialogue.

Dialogue Generation

Domain Transfer in Dialogue Systems without Turn-Level Supervision

1 code implementation16 Sep 2019 Joachim Bingel, Victor Petrén Bach Hansen, Ana Valeria Gonzalez, Paweł Budzianowski, Isabelle Augenstein, Anders Søgaard

Task oriented dialogue systems rely heavily on specialized dialogue state tracking (DST) modules for dynamically predicting user intent throughout the conversation.

Dialogue State Tracking Task-Oriented Dialogue Systems

Back to the Future -- Sequential Alignment of Text Representations

1 code implementation8 Sep 2019 Johannes Bjerva, Wouter Kouw, Isabelle Augenstein

In particular, language evolution causes data drift between time-steps in sequential decision-making tasks.

Decision Making Rumour Detection

Transductive Auxiliary Task Self-Training for Neural Multi-Task Models

no code implementations WS 2019 Johannes Bjerva, Katharina Kann, Isabelle Augenstein

Multi-task learning and self-training are two common ways to improve a machine learning model's performance in settings with limited training data.

Multi-Task Learning

X-WikiRE: A Large, Multilingual Resource for Relation Extraction as Machine Comprehension

1 code implementation WS 2019 Mostafa Abdou, Cezar Sas, Rahul Aralikatte, Isabelle Augenstein, Anders Søgaard

Although the vast majority of knowledge bases KBs are heavily biased towards English, Wikipedias do cover very different topics in different languages.

Reading Comprehension Relation Extraction

Uncovering Probabilistic Implications in Typological Knowledge Bases

no code implementations ACL 2019 Johannes Bjerva, Yova Kementchedjhieva, Ryan Cotterell, Isabelle Augenstein

The study of linguistic typology is rooted in the implications we find between linguistic features, such as the fact that languages with object-verb word ordering tend to have post-positions.

Knowledge Base Population

Issue Framing in Online Discussion Fora

no code implementations NAACL 2019 Mareike Hartmann, Tallulah Jansen, Isabelle Augenstein, Anders Søgaard

In online discussion fora, speakers often make arguments for or against something, say birth control, by highlighting certain aspects of the topic.

A Probabilistic Generative Model of Linguistic Typology

1 code implementation NAACL 2019 Johannes Bjerva, Yova Kementchedjhieva, Ryan Cotterell, Isabelle Augenstein

In the principles-and-parameters framework, the structural features of languages depend on parameters that may be toggled on or off, with a single parameter often dictating the status of multiple features.

What do Language Representations Really Represent?

no code implementations CL 2019 Johannes Bjerva, Robert Östling, Maria Han Veiga, Jörg Tiedemann, Isabelle Augenstein

If the corpus is multilingual, the same model can be used to learn distributed representations of languages, such that similar languages end up with similar representations.

Language Modelling

Copenhagen at CoNLL--SIGMORPHON 2018: Multilingual Inflection in Context with Explicit Morphosyntactic Decoding

no code implementations CONLL 2018 Yova Kementchedjhieva, Johannes Bjerva, Isabelle Augenstein

This paper documents the Team Copenhagen system which placed first in the CoNLL--SIGMORPHON 2018 shared task on universal morphological reinflection, Task 2 with an overall accuracy of 49. 87.

Morphological Inflection Multi-Task Learning

Nightmare at test time: How punctuation prevents parsers from generalizing

no code implementations WS 2018 Anders Søgaard, Miryam de Lhoneux, Isabelle Augenstein

Punctuation is a strong indicator of syntactic structure, and parsers trained on text with punctuation often rely heavily on this signal.

Parameter sharing between dependency parsers for related languages

1 code implementation EMNLP 2018 Miryam de Lhoneux, Johannes Bjerva, Isabelle Augenstein, Anders Søgaard

We find that sharing transition classifier parameters always helps, whereas the usefulness of sharing word and/or character LSTM parameters varies.

A strong baseline for question relevancy ranking

no code implementations EMNLP 2018 Ana V. González-Garduño, Isabelle Augenstein, Anders Søgaard

The best systems at the SemEval-16 and SemEval-17 community question answering shared tasks -- a task that amounts to question relevancy ranking -- involve complex pipelines and manual feature engineering.

Community Question Answering Feature Engineering

Jack the Reader -- A Machine Reading Framework

1 code implementation ACL 2018 Dirk Weissenborn, Pasquale Minervini, Isabelle Augenstein, Johannes Welbl, Tim Rockt{\"a}schel, Matko Bo{\v{s}}njak, Jeff Mitchell, Thomas Demeester, Tim Dettmers, Pontus Stenetorp, Sebastian Riedel

For example, in Question Answering, the supporting text can be newswire or Wikipedia articles; in Natural Language Inference, premises can be seen as the supporting text and hypotheses as questions.

Information Retrieval Link Prediction +4

Jack the Reader - A Machine Reading Framework

2 code implementations20 Jun 2018 Dirk Weissenborn, Pasquale Minervini, Tim Dettmers, Isabelle Augenstein, Johannes Welbl, Tim Rocktäschel, Matko Bošnjak, Jeff Mitchell, Thomas Demeester, Pontus Stenetorp, Sebastian Riedel

For example, in Question Answering, the supporting text can be newswire or Wikipedia articles; in Natural Language Inference, premises can be seen as the supporting text and hypotheses as questions.

Link Prediction Natural Language Inference +3

Multi-task Learning of Pairwise Sequence Classification Tasks Over Disparate Label Spaces

1 code implementation NAACL 2018 Isabelle Augenstein, Sebastian Ruder, Anders Søgaard

We combine multi-task learning and semi-supervised learning by inducing a joint embedding space between disparate label spaces and learning transfer functions between label embeddings, enabling us to jointly leverage unlabelled data and auxiliary, annotated datasets.

General Classification Multi-Task Learning +1

From Phonology to Syntax: Unsupervised Linguistic Typology at Different Levels with Language Embeddings

no code implementations NAACL 2018 Johannes Bjerva, Isabelle Augenstein

A core part of linguistic typology is the classification of languages according to linguistic properties, such as those detailed in the World Atlas of Language Structure (WALS).

Morphological Inflection Part-Of-Speech Tagging

Discourse-Aware Rumour Stance Classification in Social Media Using Sequential Classifiers

no code implementations6 Dec 2017 Arkaitz Zubiaga, Elena Kochkina, Maria Liakata, Rob Procter, Michal Lukasik, Kalina Bontcheva, Trevor Cohn, Isabelle Augenstein

We show that sequential classifiers that exploit the use of discourse properties in social media conversations while using only local features, outperform non-sequential classifiers.

Classification General Classification +1

Tracking Typological Traits of Uralic Languages in Distributed Language Representations

no code implementations WS 2018 Johannes Bjerva, Isabelle Augenstein

Although linguistic typology has a long history, computational approaches have only recently gained popularity.

Latent Multi-task Architecture Learning

2 code implementations23 May 2017 Sebastian Ruder, Joachim Bingel, Isabelle Augenstein, Anders Søgaard

In practice, however, MTL involves searching an enormous space of possible parameter sharing architectures to find (a) the layers or subspaces that benefit from sharing, (b) the appropriate amount of sharing, and (c) the appropriate relative weights of the different task losses.

Multi-Task Learning

Turing at SemEval-2017 Task 8: Sequential Approach to Rumour Stance Classification with Branch-LSTM

1 code implementation SEMEVAL 2017 Elena Kochkina, Maria Liakata, Isabelle Augenstein

This paper describes team Turing's submission to SemEval 2017 RumourEval: Determining rumour veracity and support for rumours (SemEval 2017 Task 8, Subtask A).

Classification General Classification +3

SemEval 2017 Task 10: ScienceIE - Extracting Keyphrases and Relations from Scientific Publications

1 code implementation SEMEVAL 2017 Isabelle Augenstein, Mrinal Das, Sebastian Riedel, Lakshmi Vikraman, Andrew McCallum

We describe the SemEval task of extracting keyphrases and relations between them from scientific documents, which is crucial for understanding which publications describe which processes, tasks and materials.

Knowledge Base Population

Multi-Task Learning of Keyphrase Boundary Classification

no code implementations ACL 2017 Isabelle Augenstein, Anders Søgaard

Keyphrase boundary classification (KBC) is the task of detecting keyphrases in scientific articles and labelling them with respect to predefined types.

Classification General Classification +1

Generalisation in Named Entity Recognition: A Quantitative Analysis

no code implementations11 Jan 2017 Isabelle Augenstein, Leon Derczynski, Kalina Bontcheva

Unseen NEs, in particular, play an important role, which have a higher incidence in diverse genres such as social media than in more regular genres such as newswire.

Named Entity Recognition NER

emoji2vec: Learning Emoji Representations from their Description

7 code implementations WS 2016 Ben Eisner, Tim Rocktäschel, Isabelle Augenstein, Matko Bošnjak, Sebastian Riedel

Many current natural language processing applications for social media rely on representation learning and utilize pre-trained word embeddings.

Sentiment Analysis Word Embeddings

Numerically Grounded Language Models for Semantic Error Correction

no code implementations EMNLP 2016 Georgios P. Spithourakis, Isabelle Augenstein, Sebastian Riedel

Semantic error detection and correction is an important task for applications such as fact checking, speech-to-text or grammatical error correction.

Fact Checking Grammatical Error Correction +1

Stance Detection with Bidirectional Conditional Encoding

1 code implementation EMNLP 2016 Isabelle Augenstein, Tim Rocktäschel, Andreas Vlachos, Kalina Bontcheva

Stance detection is the task of classifying the attitude expressed in a text towards a target such as Hillary Clinton to be "positive", negative" or "neutral".

Stance Detection

Monolingual Social Media Datasets for Detecting Contradiction and Entailment

no code implementations LREC 2016 Piroska Lendvai, Isabelle Augenstein, Kalina Bontcheva, Thierry Declerck

Entailment recognition approaches are useful for application domains such as information extraction, question answering or summarisation, for which evidence from multiple sentences needs to be combined.

Natural Language Inference Question Answering

USFD: Twitter NER with Drift Compensation and Linked Data

no code implementations WS 2015 Leon Derczynski, Isabelle Augenstein, Kalina Bontcheva

This paper describes a pilot NER system for Twitter, comprising the USFD system entry to the W-NUT 2015 NER shared task.

NER

Cannot find the paper you are looking for? You can Submit a new open access paper.