Abdelrahman
Mohamed

My research focuses on multimodal learning, particularly emotion understanding and affective reasoning. I am interested in how models combine information from speech, video, and language, and which cues they rely on when interpreting emotions. I am especially interested in situations where these cues are subtle, ambiguous, or conflicting, and understanding how models handle these situations and which cues influence their predictions..

I am also interested in how cultural differences shape emotional expression and perception, and how these differences affect emotion recognition across cultures. My work also explores temporal emotion understanding, cross-modal alignment, and model interpretability.