Causal Intersectionality and Dual Form of Gradient Descent for Multimodal Analysis: a Case Study on Hateful Memes

1 code implementation • 19 Aug 2023 • Yosuke Miyanishi, Minh Le Nguyen

This paper demonstrates that hateful meme detection can be viewed as an ATE estimation using intersectionality principles, and summarized gradient-based attention scores highlight distinct behaviors of three Transformer models.

Decision Making In-Context Learning