mlsanityData to 7 Oct 2026

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

arXiv 2209.08681 ↗First posted 18 Sep 2022cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

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 source-specific information that negatively impacts its domain invariance. Prior work has attempted to penalize terms related to hate-speech 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 hate-speech classes, yielding consistent improvements in cross-domain evaluation.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 58 references that are in the map. First posted on arXiv 18 Sep 2022.

Builds on

29 of 58 references.

References · 29

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  4. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  5. 2017Adversarial Discriminative Domain Adaptationcited by 1,309
  6. 2019Is Attention Interpretable?cited by 387
  7. 2020Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detectioncited by 217
  8. 2020Debugging Tests for Model Explanationscited by 146
  9. 2019Interpretations are useful: penalizing explanations to align neural networks with prior knowledgecited by 144
  10. 2019A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Mediacited by 95
  11. 2020Contextualizing Hate Speech Classifiers with Post-hoc Explanationcited by 84
  12. 2019Incorporating Priors with Feature Attribution on Text Classificationcited by 68
17 more
  1. 2021Towards generalisable hate speech detection: a review on obstacles and solutionscited by 66
  2. 2022Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Listscited by 54
  3. 2020PERL: Pivot-based Domain Adaptation for Pre-trained Deep Contextualized Embedding Modelscited by 33
  4. 2021Improving the Faithfulness of Attention-based Explanations with Task-specific Information for Text Classificationcited by 29
  5. 2020Explainable Deep Classification Models for Domain Generalizationcited by 27
  6. 2021Refining Language Models with Compositional Explanationscited by 27
  7. 2019Towards Hierarchical Importance Attribution: Explaining Compositional Semantics for Neural Sequence Modelscited by 26
  8. 2018Challenges in Discriminating Profanity from Hate Speechcited by 23
  9. 2022An Empirical Study on Explanations in Out-of-Domain Settingscited by 21
  10. 2020Knowledge Distillation for BERT Unsupervised Domain Adaptationcited by 14
  11. 2020Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysiscited by 10
  12. 2021Flexible Instance-Specific Rationalization of NLP Modelscited by 10
  13. 2019Towards non-toxic landscapes: Automatic toxic comment detection using DNNcited by 8
  14. 2020Adversarial Augmentation Policy Search for Domain and Cross-Lingual Generalization in Reading Comprehensioncited by 7
  15. 2021Unsupervised Domain Adaptation for Hate Speech Detection Using a Data Augmentation Approachcited by 6
  16. 2022Identifying and Characterizing Active Citizens who Refute Misinformation in Social Mediacited by 5
  17. 2022Dynamically Refined Regularization for Improving Cross-corpora Hate Speech Detectioncited by 3

Built on by

Not cited yet.