Dynamically Refined Regularization for Improving Cross-corpora Hate Speech Detection

Bose, T.; Αλέτρας, Νικόλαος; Illina, Irina; Fohr, Dominique

doi:10.18653/v1/2022.findings-acl.32

Cited by 4 publications

(2 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current HS models often suffer from spurious correlations (Ramponi and Tonelli, 2022;Bose et al, 2022;Wang et al, 2022): they tend to utilize prediction patterns that hold for the majority examples but do not hold in general. This might cause the model to be biased in applications (Ramponi and Tonelli, 2022;Gardner et al, 2021).…”

Section: Identifying Spurious Correlationsmentioning

confidence: 99%

Mitigating Biases in Hate Speech Detection from A Causal Perspective

Zhang,

Chen,

Yang

2023

Findings of the Association for Computational Linguistics: EMNLP 2023

View full text Add to dashboard Cite

Warning: This paper discusses and contains offensive or upsetting content. Nowadays, many hate speech detectors are built to automatically detect hateful content. However, their training sets are sometimes skewed towards certain stereotypes (e.g., race or religion-related). As a result, the detectors are prone to depend on some shortcuts for predictions. Previous works mainly focus on token-level analysis and heavily rely on human experts' annotations to identify spurious correlations, which is not only costly but also incapable of discovering higherlevel artifacts. In this work, we use grammar induction to find grammar patterns for hate speech and analyze this phenomenon from a causal perspective. Concretely, we categorize and verify different biases based on their spuriousness and influence on the model prediction. Then, we propose two mitigation approaches including Multi-Task Intervention and Data-Specific Intervention based on these confounders. Experiments conducted on 9 hate speech datasets demonstrate the effectiveness of our approaches. The code is available at https://github. com/SALT-NLP/Bias_Hate_Causal.

show abstract

Section: Identifying Spurious Correlationsmentioning

confidence: 99%

Mitigating Biases in Hate Speech Detection from A Causal Perspective

Zhang,

Chen,

Yang

2023

Findings of the Association for Computational Linguistics: EMNLP 2023

View full text Add to dashboard Cite

show abstract

“…Chrysostomou and Aletras (2022a) further show that post-hoc explanation methods might not provide faithful explanations in out-of-domain settings. The contemporaneous work by Attanasio et al (2022) and Bose et al (2022) reduce lexical overfitting automatically with entropy-based attentions and feature attributions, respectively. While cross-domain classification performance across different datasets is not studied in the former, the latter needs some labeled target instances to identify the over-fitted terms.…”

Section: Introductionmentioning

confidence: 99%

Domain Classification-based Source-specific Term Penalization for Domain Adaptation in Hate-speech Detection

Bose¹,

Αλέτρας²,

Illina³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Warning: this paper contains content that may be offensive and distressing.State-of-the-art approaches for hate-speech detection usually exhibit poor performance in out-of-domain settings. This occurs, typically, due to classifiers overemphasizing sourcespecific information that negatively impacts its domain invariance. Prior work has attempted to penalize terms related to hatespeech from manually curated lists using feature attribution methods, which quantify the importance assigned to input terms by the classifier when making a prediction. We, instead, propose a domain adaptation approach that automatically extracts and penalizes source-specific terms using a domain classifier, which learns to differentiate between domains, and feature-attribution scores for hatespeech classes, yielding consistent improvements in cross-domain evaluation.

show abstract

Exploration of Multi-corpus Learning for Hate Speech Classification in Low Resource Scenarios

D'Sa

Illina

Fohr

et al. 2022

Text, Speech, and Dialogue

Self Cite

View full text Add to dashboard Cite

The dramatic increase in social media has given rise to the problem of online hate speech. Deep neural network-based classifiers have become the state-of-the-art for automatic hate speech classification. The performance of these classifiers depends on the amount of available labelled training data. However, most hate speech corpora have a small number of hate speech samples. In this article, we aim to jointly use multiple hate speech corpora to improve hate speech classification performance in low-resource scenarios. We harness different hate speech corpora in a multi-task learning setup by associating one task to one corpus. This multi-corpus learning scheme is expected to improve the generalization, the latent representations, and domain adaptation of the model. Our work evaluates multi-corpus learning for hate speech classification and domain adaptation. We show significant improvements in classification and domain adaptation in low-resource scenarios.

show abstract

Dynamically Refined Regularization for Improving Cross-corpora Hate Speech Detection

Cited by 4 publications

References 18 publications

Mitigating Biases in Hate Speech Detection from A Causal Perspective

Mitigating Biases in Hate Speech Detection from A Causal Perspective

Domain Classification-based Source-specific Term Penalization for Domain Adaptation in Hate-speech Detection

Exploration of Multi-corpus Learning for Hate Speech Classification in Low Resource Scenarios

Contact Info

Product

Resources

About