# Improving Generalizability in Implicitly Abusive Language Detection with Concept Activation Vectors

arXiv: https://arxiv.org/abs/2204.02261 · HTML: https://mlsanity.com/paper/2204.02261/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 41 references that are in the map.
- First posted on arXiv 5 Apr 2022.

## Builds on (most cited first, 17 of 17)

- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Social Bias Frames: Reasoning about Social and Power Implications of Language](https://mlsanity.com/paper/1911.03891/) (2019)
- [Pretrained Transformers Improve Out-of-Distribution Robustness](https://mlsanity.com/paper/2004.06100/) (2020)
- [Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior](https://mlsanity.com/paper/1802.00393/) (2018)
- [Ex Machina: Personal Attacks Seen at Scale](https://mlsanity.com/paper/1610.08914/) (2016)
- [Directions in Abusive Language Training Data: Garbage In, Garbage Out](https://mlsanity.com/paper/2004.01670/) (2020)
- [Understanding Abuse: A Typology of Abusive Language Detection Subtasks](https://mlsanity.com/paper/1705.09899/) (2017)
- [Challenges in Automated Debiasing for Toxic Language Detection](https://mlsanity.com/paper/2102.00086/) (2021)
- [Towards generalisable hate speech detection: a review on obstacles and solutions](https://mlsanity.com/paper/2102.08886/) (2021)
- [A Survey of Active Learning for Text Classification using Deep Neural Networks](https://mlsanity.com/paper/2008.07267/) (2020)
- [Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective](https://mlsanity.com/paper/2012.12305/) (2020)
- [Stereotypical Bias Removal for Hate Speech Detection Task using Knowledge-based Generalizations](https://mlsanity.com/paper/2001.05495/) (2020)
- [Detecting East Asian Prejudice on Social Media](https://mlsanity.com/paper/2005.03909/) (2020)
- [Unsupervised Discovery of Implicit Gender Bias](https://mlsanity.com/paper/2004.08361/) (2020)
- [Towards Linguistically Generalizable NLP Systems: A Workshop and Shared Task](https://mlsanity.com/paper/1711.01505/) (2017)
- [On Cross-Dataset Generalization in Automatic Detection of Online Abuse](https://mlsanity.com/paper/2010.07414/) (2020)
- [Active Learning for Domain Classification in a Commercial Spoken Personal Assistant](https://mlsanity.com/paper/1908.11404/) (2019)

## Built on by (most cited first, 14 of 14)

- [Challenges in Applying Explainability Methods to Improve the Fairness of NLP Models](https://mlsanity.com/paper/2206.03945/) (2022)
- [From Neural Activations to Concepts: A Survey on Explaining Concepts in Neural Networks](https://mlsanity.com/paper/2310.11884/) (2023)
- [LG-CAV: Train Any Concept Activation Vector with Language Guidance](https://mlsanity.com/paper/2410.10308/) (2024)
- [Towards Achieving Concept Completeness for Textual Concept Bottleneck Models](https://mlsanity.com/paper/2502.11100/) (2025)
- [Towards Procedural Fairness: Uncovering Biases in How a Toxic Language Classifier Uses Sentiment Information](https://mlsanity.com/paper/2210.10689/) (2022)
- [K-HATERS: A Hate Speech Detection Corpus in Korean with Target-Specific Ratings](https://mlsanity.com/paper/2310.15439/) (2023)
- [Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers](https://mlsanity.com/paper/2307.01900/) (2023)
- [Explaining Deep Neural Networks for Bearing Fault Detection with Vibration Concepts](https://mlsanity.com/paper/2310.11450/) (2023)
- [KTCR: Improving Implicit Hate Detection with Knowledge Transfer driven Concept Refinement](https://mlsanity.com/paper/2410.15314/) (2024)
- [Freeze and Reveal: Exposing Modality Bias in Vision-Language Models](https://mlsanity.com/paper/2508.07432/) (2025)
- [Estimating Semantic Similarity between In-Domain and Out-of-Domain Samples](https://mlsanity.com/paper/2306.01206/) (2023)
- [Improving Cross-Domain Hate Speech Generalizability with Emotion Knowledge](https://mlsanity.com/paper/2311.14865/) (2023)
- [Concept-Based Interpretability for Toxicity Detection](https://mlsanity.com/paper/2511.16689/) (2025)
- [Analyzing Traditional and Neural Approaches to Multilingual Readability Assessment](https://mlsanity.com/paper/2609.10792/) (2026)

## Abstract (arXiv)

Robustness of machine learning models on ever-changing real-world data is critical, especially for applications affecting human well-being such as content moderation. New kinds of abusive language continually emerge in online discussions in response to current events (e.g., COVID-19), and the deployed abuse detection systems should be updated regularly to remain accurate. In this paper, we show that general abusive language classifiers tend to be fairly reliable in detecting out-of-domain explicitly abusive utterances but fail to detect new types of more subtle, implicit abuse. Next, we propose an interpretability technique, based on the Testing Concept Activation Vector (TCAV) method from computer vision, to quantify the sensitivity of a trained model to the human-defined concepts of explicit and implicit abusive language, and use that to explain the generalizability of the model on new data, in this case, COVID-related anti-Asian hate speech. Extending this technique, we introduce a novel metric, Degree of Explicitness, for a single instance and show that the new metric is beneficial in suggesting out-of-domain unlabeled examples to effectively enrich the training data with informative, implicitly abusive texts.