Bridging the Gaps: Multi Task Learning for Domain Transfer of Hate Speech Detection

By Zeerak Talat, James Thorne, Joachim Bingel. 2018

93 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

HateCheck: Functional Tests for Hate Speech Detection Models (2020, background): “…in data source, sampling strategy and annotation process (Vidgen and Derczynski, 2020; Poletto et al., 2020), and are known to exhibit annotator biases (Waseem, 2016; Waseem et al., 2018; Sap et al., 2019) as well as topic and author biases (Wiegand et al., 2019; Nejadgholi and Kiritchenko, 2020).”
Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection (2021, background): “Similarly, many models overfit on the use of slurs and pejorative terms, treating them as hateful irrespective of how they are used (Waseem et al., 2018; Davidson et al., 2017; Kurrek et al., 2020; Palmer et al., 2020).”
Racial Bias in Hate Speech and Abusive Language Detection Datasets (2019, methodology): “First, the dataset contains a large number of cases where AAE is used ( Waseem et al., 2018).”
Disembodied Machine Learning: On the Illusion of Objectivity in NLP (2021, background): “Waseem, Thorne, and Bingel (2018) argue that a consequence of embodying a white perspective of respectability is that almost all instances of the n-word are tagged as the positive classes.”
Misogyny Detection in Twitter: a Multilingual and Cross-Domain Study (2020, unknown): “In Waseem et al. (2018) the first attempt to deal with cross-domain classification in an abusive language detection task is reported, by proposing a multi-tasks learning (MTL) approach.”
A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media (2019, methodology): “Waseem et al. [23] brought a new insight to hate speech and abusive language detection tasks by proposing a multi-task learning framework to deal with datasets across diﬀerent annotation schemes, labels, or geographic and cultural inﬂuences from data sampling.”
Towards Ethics by Design in Online Abusive Content Detection (2020, background): “Similarly, annotators’ insensitivity or unawareness of dialect can lead to biased annotations and amplify harms against racial minorities (Waseem et al., 2018; Sap et al., 2019).”
Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective (2020, background): “Waseem, Thorne, and Bingel (2018) demonstrated that learning to detect hate speech alongside an auxiliary task improved robustness across datasets originating from different distributions and labeled under differing annotation guidelines.”
Studying Generalisability across Abusive Language Detection Datasets (2019, background): “Similarly, Waseem et al. (2018) attempted to address the problem of differences between datasets by building a robust multi-task learning model, which improves upon single-task performance by using auxiliary samples from select datasets.”
Tackling Online Abuse: A Survey of Automated Abuse Detection Methods (2019, background): “Recently, Waseem et al. (2018) and Karan and ˇSnajder (2018) exploited multi-task learning frameworks to train models that are robust across data from different distributions and data annotated under different guidelines.”
Cross-Domain Detection of Abusive Language Online (2018, background): “The work closest to ours is (Waseem et al., 2018), where multi-task learning is used to build robust hate-speech detection models.”
Towards multidomain and multilingual abusive language detection: a survey (2021, methodology): “In addition, Waseem et al. [145] proposed a multitask learning architecture based on LSTM to learn the problem from two or more tasks sequentially, providing a medium for knowledge transfer between domains.”
XHate-999: Analyzing and Detecting Abusive Language Across Domains and Languages (2020, unknown): “Another line of research (Wiegand et al., 2018a; Karan and Šnajder, 2018; Waseem et al., 2018, inter alia ) focuses on beneﬁts of cross-domain transfer in monolingual settings.”