Hateful Meme Detection Through Context-Sensitive Prompting and Fine-Grained Labeling (Student Abstract)

Rongxin Ouyang, Kokil Jaidka, Subhayan Mukerjee, Guangyu Cui

AAAI 2025 pp. 29459-29461

doi:10.1609/AAAI.V39I28.35287 /aaai/2025/ouyang2025aaai-hateful/

Abstract

The prevalence of multi-modal content on social media complicates automated moderation strategies. This calls for an enhancement in multi-modal classification and a deeper understanding of understated meanings in images and memes. Although previous efforts have aimed at improving model performance through fine-tuning, few have explored an end-to-end optimization pipeline that accounts for modalities, prompting, labelling, and fine-tuning. In this study, we propose an end-to-end conceptual framework for model opti- mization in complex tasks. Experiments support the efficacy of this traditional yet novel framework, achieving the highest accuracy and AUROC. Ablation experiments demonstrate that isolated optimisations are not ineffective on their own.

AAAI Semantic Scholar

Cite

Text

Ouyang et al. "Hateful Meme Detection Through Context-Sensitive Prompting and Fine-Grained Labeling (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I28.35287

Markdown

[Ouyang et al. "Hateful Meme Detection Through Context-Sensitive Prompting and Fine-Grained Labeling (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/ouyang2025aaai-hateful/) doi:10.1609/AAAI.V39I28.35287

BibTeX

@inproceedings{ouyang2025aaai-hateful,
  title     = {{Hateful Meme Detection Through Context-Sensitive Prompting and Fine-Grained Labeling (Student Abstract)}},
  author    = {Ouyang, Rongxin and Jaidka, Kokil and Mukerjee, Subhayan and Cui, Guangyu},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2025},
  pages     = {29459-29461},
  doi       = {10.1609/AAAI.V39I28.35287},
  url       = {https://mlanthology.org/aaai/2025/ouyang2025aaai-hateful/}
}