# Deciphering Implicit Hate: Evaluating Automated Detection Algorithms for Multimodal Hate

arXiv: https://arxiv.org/abs/2106.05903 · HTML: https://mlsanity.com/paper/2106.05903/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 82 references that are in the map.
- First posted on arXiv 10 Jun 2021.

## Builds on (most cited first, 30 of 33)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning](https://mlsanity.com/paper/1602.07261/) (2016)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [Learning Transferable Architectures for Scalable Image Recognition](https://mlsanity.com/paper/1707.07012/) (2017)
- [Bag of Tricks for Efficient Text Classification](https://mlsanity.com/paper/1607.01759/) (2016)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [12-in-1: Multi-Task Vision and Language Representation Learning](https://mlsanity.com/paper/1912.02315/) (2019)
- [Ex Machina: Personal Attacks Seen at Scale](https://mlsanity.com/paper/1610.08914/) (2016)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval)](https://mlsanity.com/paper/1903.08983/) (2019)
- [Deep Learning for Hate Speech Detection in Tweets](https://mlsanity.com/paper/1706.00188/) (2017)
- [Supervised Multimodal Bitransformers for Classifying Images and Text](https://mlsanity.com/paper/1909.02950/) (2019)
- [Understanding Abuse: A Typology of Abusive Language Detection Subtasks](https://mlsanity.com/paper/1705.09899/) (2017)
- [Exploring Hate Speech Detection in Multimodal Publications](https://mlsanity.com/paper/1910.03814/) (2019)
- [A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media](https://mlsanity.com/paper/1910.12574/) (2019)
- [Contextualizing Hate Speech Classifiers with Post-hoc Explanation](https://mlsanity.com/paper/2005.02439/) (2020)
- [Detecting Online Hate Speech Using Context Aware Models](https://mlsanity.com/paper/1710.07395/) (2017)
- [Challenges for Toxic Comment Classification: An In-Depth Error Analysis](https://mlsanity.com/paper/1809.07572/) (2018)
- [Universal Language Model Fine-tuning for Text Classification](https://mlsanity.com/paper/1801.06146/) (2018)
- [On the Origins of Memes by Means of Fringe Web Communities](https://mlsanity.com/paper/1805.12512/) (2018)
- [Detecting East Asian Prejudice on Social Media](https://mlsanity.com/paper/2005.03909/) (2020)
- [Hate Speech in Pixels: Detection of Offensive Memes towards Automatic Moderation](https://mlsanity.com/paper/1910.02334/) (2019)
- [Leveraging Intra-User and Inter-User Representation Learning for Automated Hate Speech Detection](https://mlsanity.com/paper/1804.03124/) (2018)
- [Demographic Inference and Representative Population Estimates from Multilingual Social Media Data](https://mlsanity.com/paper/1905.05961/) (2019)
- [Detecting Offensive Language in Tweets Using Deep Learning](https://mlsanity.com/paper/1801.04433/) (2018)

## Built on by (most cited first, 8 of 8)

- [From Dogwhistles to Bullhorns: Unveiling Coded Rhetoric with Language Models](https://mlsanity.com/paper/2305.17174/) (2023)
- [HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter](https://mlsanity.com/paper/2411.15462/) (2024)
- [The State of Profanity Obfuscation in Natural Language Processing](https://mlsanity.com/paper/2210.07595/) (2022)
- [Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification](https://mlsanity.com/paper/2303.15016/) (2023)
- [Improving Multimodal Classification of Social Media Posts by Leveraging Image-Text Auxiliary Tasks](https://mlsanity.com/paper/2309.07794/) (2023)
- [Focal Inferential Infusion Coupled with Tractable Density Discrimination for Implicit Hate Detection](https://mlsanity.com/paper/2309.11896/) (2023)
- [Multilingual Hate Speech Detection and Counterspeech Generation: A Comprehensive Survey and Practical Guide](https://mlsanity.com/paper/2603.19279/) (2026)
- [Towards Training-free Multimodal Hate Localisation with Large Language Models](https://mlsanity.com/paper/2602.09637/) (2026)

## Abstract (arXiv)

Accurate detection and classification of online hate is a difficult task. Implicit hate is particularly challenging as such content tends to have unusual syntax, polysemic words, and fewer markers of prejudice (e.g., slurs). This problem is heightened with multimodal content, such as memes (combinations of text and images), as they are often harder to decipher than unimodal content (e.g., text alone). This paper evaluates the role of semantic and multimodal context for detecting implicit and explicit hate. We show that both text- and visual- enrichment improves model performance, with the multimodal model (0.771) outperforming other models' F1 scores (0.544, 0.737, and 0.754). While the unimodal-text context-aware (transformer) model was the most accurate on the subtask of implicit hate detection, the multimodal model outperformed it overall because of a lower propensity towards false positives. We find that all models perform better on content with full annotator agreement and that multimodal models are best at classifying the content where annotators disagree. To conduct these investigations, we undertook high-quality annotation of a sample of 5,000 multimodal entries. Tweets were annotated for primary category, modality, and strategy. We make this corpus, along with the codebook, code, and final model, freely available.