# TRACE: Textual Relevance Augmentation and Contextual Encoding for Multimodal Hate Detection

arXiv: https://arxiv.org/abs/2504.17902 · HTML: https://mlsanity.com/paper/2504.17902/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 61 references that are in the map.
- First posted on arXiv 24 Apr 2025.

## Builds on (most cited first, 22 of 22)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes](https://mlsanity.com/paper/2005.04790/) (2020)
- [YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information](https://mlsanity.com/paper/2402.13616/) (2024)
- [Recognize Anything: A Strong Image Tagging Model](https://mlsanity.com/paper/2306.03514/) (2023)
- [Exploring Hate Speech Detection in Multimodal Publications](https://mlsanity.com/paper/1910.03814/) (2019)
- [Is Cosine-Similarity of Embeddings Really About Similarity?](https://mlsanity.com/paper/2403.05440/) (2024)
- [Hate-CLIPper: Multimodal Hateful Meme Classification based on Cross-modal Interaction of CLIP Features](https://mlsanity.com/paper/2210.05916/) (2022)
- [ViGoR: Improving Visual Grounding of Large Vision Language Models with Fine-Grained Reward Modeling](https://mlsanity.com/paper/2402.06118/) (2024)
- [Learning Visual Grounding from Generative Vision and Language Model](https://mlsanity.com/paper/2407.14563/) (2024)
- [Improving Visual Grounding by Encouraging Consistent Gradient-based Explanations](https://mlsanity.com/paper/2206.15462/) (2022)
- [Towards a Robust Framework for Multimodal Hate Detection: A Study on Video vs. Image-based Content](https://mlsanity.com/paper/2502.07138/) (2025)
- [Tell Codec What Worth Compressing: Semantically Disentangled Image Coding for Machine with LMMs](https://mlsanity.com/paper/2408.08575/) (2024)

## Built on by (most cited first, 3 of 3)

- [Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification](https://mlsanity.com/paper/2609.26907/) (2026)
- [Labels or Input? Rethinking Augmentation in Multimodal Hate Detection](https://mlsanity.com/paper/2508.11808/) (2025)
- [Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection](https://mlsanity.com/paper/2609.18860/) (2026)

## Abstract (arXiv)

Social media memes are a challenging domain for hate detection because they intertwine visual and textual cues into culturally nuanced messages. To tackle these challenges, we introduce TRACE, a hierarchical multimodal framework that leverages visually grounded context augmentation, along with a novel caption-scoring network to emphasize hate-relevant content, and parameter-efficient fine-tuning of CLIP's text encoder. Our experiments demonstrate that selectively fine-tuning deeper text encoder layers significantly enhances performance compared to simpler projection-layer fine-tuning methods. Specifically, our framework achieves state-of-the-art accuracy (0.807) and F1-score (0.806) on the widely-used Hateful Memes dataset, matching the performance of considerably larger models while maintaining efficiency. Moreover, it achieves superior generalization on the MultiOFF offensive meme dataset (F1-score 0.673), highlighting robustness across meme categories. Additional analyses confirm that robust visual grounding and nuanced text representations significantly reduce errors caused by benign confounders. We publicly release our code to facilitate future research.