Adapting Reinforcement Learning with Chain-of-Thought Supervision for Explainable Detection of Hateful and Propagandistic Memes

Mohamed Bayan Kmainasi, Mucahid Kutlu, Ali Ezzat Shahroor, Abul Hasnat, Firoj AlamarXiv 2026 · DOI

BibTeX
@misc{kmainasi2026adapting,
  author = {Mohamed Bayan Kmainasi and Mucahid Kutlu and Ali Ezzat Shahroor and Abul Hasnat and Firoj Alam},
  title = {{Adapting Reinforcement Learning with Chain-of-Thought Supervision for Explainable Detection of Hateful and Propagandistic Memes}},
  year = {2026},
  eprint = {2606.15307},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.15307},
}

Loading related content...

Figure