S1 E1 Approaching VisualQA
Train your first V+L model
S1 E2 Gated Fusion Networks for Modality Attention
Learn to turn and off V+L modalities.
S1 E3a Visualizing a Multimodal Embedding Space
Visualise V+L joint embedding space.
E1 Diagnosing and De-biasing Large Language Models
Debias a language model.