# Unveiling Covert Toxicity in Multimodal Data via Toxicity Association Graphs: A Graph-Based Metric and Interpretable Detection Framework

arXiv: https://arxiv.org/abs/2602.03268 · HTML: https://mlsanity.com/paper/2602.03268/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 44 references that are in the map.
- First posted on arXiv 3 Feb 2026.

## Builds on (most cited first, 25 of 25)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [MOMENTA: A Multimodal Framework for Detecting Harmful Memes and Their Targets](https://mlsanity.com/paper/2109.05184/) (2021)
- [Detecting Harmful Memes and Their Targets](https://mlsanity.com/paper/2110.00413/) (2021)
- [VLSBench: Unveiling Visual Leakage in Multimodal Safety](https://mlsanity.com/paper/2411.19939/) (2024)
- [Hate-CLIPper: Multimodal Hateful Meme Classification based on Cross-modal Interaction of CLIP Features](https://mlsanity.com/paper/2210.05916/) (2022)
- [GOAT-Bench: Safety Insights to Large Multimodal Models through Meme-Based Social Abuse](https://mlsanity.com/paper/2401.01523/) (2024)
- [Vilio: State-of-the-art Visio-Linguistic Models applied to Hateful Memes](https://mlsanity.com/paper/2012.07788/) (2020)
- [Detecting Hate Speech in Memes Using Multimodal Deep Learning Approaches: Prize-winning solution to Hateful Memes Challenge](https://mlsanity.com/paper/2012.12975/) (2020)
- [ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors](https://mlsanity.com/paper/2402.16444/) (2024)
- [Detecting Propaganda Techniques in Memes](https://mlsanity.com/paper/2109.08013/) (2021)
- [MemeCLIP: Leveraging CLIP Representations for Multimodal Meme Classification](https://mlsanity.com/paper/2409.14703/) (2024)
- [DISARM: Detecting the Victims Targeted by Harmful Memes](https://mlsanity.com/paper/2205.05738/) (2022)
- [DisinfoMeme: A Multimodal Dataset for Detecting Meme Intentionally Spreading Out Disinformation](https://mlsanity.com/paper/2205.12617/) (2022)
- [Meme-ingful Analysis: Enhanced Understanding of Cyberbullying in Memes Through Multimodal Explanations](https://mlsanity.com/paper/2401.09899/) (2024)
- [Towards Comprehensive Detection of Chinese Harmful Memes](https://mlsanity.com/paper/2410.02378/) (2024)
- [ToxVidLM: A Multimodal Framework for Toxicity Detection in Code-Mixed Videos](https://mlsanity.com/paper/2405.20628/) (2024)
- [Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features](https://mlsanity.com/paper/2504.00557/) (2025)
- [Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content](https://mlsanity.com/paper/2106.08409/) (2021)
- [A Template Is All You Meme](https://mlsanity.com/paper/2311.06649/) (2023)
- [ShieldVLM: Safeguarding the Multimodal Implicit Toxicity via Deliberative Reasoning with LVLMs](https://mlsanity.com/paper/2505.14035/) (2025)
- [MemeTector: Enforcing deep focus for meme detection](https://mlsanity.com/paper/2205.13268/) (2022)
- [Text or Image? What is More Important in Cross-Domain Generalization Capabilities of Hate Meme Detection Models?](https://mlsanity.com/paper/2402.04967/) (2024)

## Abstract (arXiv)

Detecting toxicity in multimodal data remains a significant challenge, as harmful meanings often lurk beneath seemingly benign individual modalities: only emerging when modalities are combined and semantic associations are activated. To address this, we propose a novel detection framework based on Toxicity Association Graphs (TAGs), which systematically model semantic associations between innocuous entities and latent toxic implications. Leveraging TAGs, we introduce the first quantifiable metric for hidden toxicity, the Multimodal Toxicity Covertness (MTC), which measures the degree of concealment in toxic multimodal expressions. By integrating our detection framework with the MTC metric, our approach enables precise identification of covert toxicity while preserving full interpretability of the decision-making process, significantly enhancing transparency in multimodal toxicity detection. To validate our method, we construct the Covert Toxic Dataset, the first benchmark specifically designed to capture high-covertness toxic multimodal instances. This dataset encodes nuanced cross-modal associations and serves as a rigorous testbed for evaluating both the proposed metric and detection framework. Extensive experiments demonstrate that our approach outperforms existing methods across both low- and high-covertness toxicity regimes, while delivering clear, interpretable, and auditable detection outcomes. Together, our contributions advance the state of the art in explainable multimodal toxicity detection and lay the foundation for future context-aware and interpretable approaches. Content Warning: This paper contains examples of toxic multimodal content that may be offensive or disturbing to some readers. Reader discretion is advised.