Steffen Eger
NLP, Deep Learning
The rapid growth of information in the field of Generative Artificial
In...
While summarization has been extensively researched in natural language
...
Available corpora for Argument Mining differ along several axes, and one...
Protecting privacy in contemporary NLP models is gaining in importance. ...
ChatGPT, a chatbot developed by OpenAI, has gained widespread popularity...
We consider the end-to-end abstract-to-title generation problem, explori...
State-of-the-art poetry generation systems are often complex. They eithe...
State-of-the-art machine translation evaluation metrics are based on
bla...
We measure support with women and migrants in German political debates o...
We explore efficient evaluation metrics for Natural Language Generation
...
The evaluation of recent embedding-based evaluation metrics for text
gen...
Recently proposed BERT-based evaluation metrics perform well on standard...
Reproducibility is of utmost concern in machine learning and natural lan...
In this paper, we classify scientific articles in the domain of natural
...
The vast majority of evaluation metrics for machine translation are
supe...
Multilingual representations pre-trained with monolingual data exhibit
c...
Recently, there has been a growing interest in designing text generation...
Evaluation in NLP is usually done by comparing the scores of competing
s...
Evaluation metrics are a key ingredient for progress of text generation
...
We analyze bias in historical corpora as encoded in diachronic distribut...
We introduce the well-established social scientific concept of social
so...
Adversarial attacks expose important blind spots of deep learning system...
We probe the layers in multilingual BERT (mBERT) for phylogenetic and
ge...
We introspect black-box sentence embeddings by conditionally generating ...
Adversarial attacks are label-preserving modifications to inputs of mach...
Multilingual representations have the potential to make cross-lingual sy...
Sentence encoders map sentences to real valued vectors for use in downst...
Most approaches to emotion analysis regarding social media, literature, ...
Due to its semantic succinctness and novelty of expression, poetry is a ...
This paper describes DBPal, a new system to translate natural language
u...
A robust evaluation metric has a profound impact on the development of t...
Obstacles hindering the development of capsule networks for challenging ...
Deep learning models continuously break new records across different NLP...
Visual modifications to text are often used to obfuscate offensive comme...
Peer review is a core element of the scientific process, particularly in...
We perform trend detection on two datasets of Arxiv papers, derived from...
Activation functions play a crucial role in neural networks because they...
Argumentation mining (AM) requires the identification of complex discour...
We investigate whether and where multi-task learning (MTL) can improve
p...
We investigate whether and where multi-task learning (MTL) can improve
p...
Average word embeddings are a common baseline for more sophisticated sen...
Argument mining has become a popular research area in NLP. It typically
...
We investigate neural techniques for end-to-end computational argumentat...
We consider two graph models of semantic change. The first is a time-ser...
This paper describes our approach to the SemEval 2017 Task 10: "Extracti...
We analyze the performance of encoder-decoder neural models and compare ...
We study the role of the second language in bilingual word embeddings in...
We count the number of alignments of N > 1 sequences when match-up types...