Multimodality and Deep Learning
Tim explores the concept of multimodality in understanding robust object recognition, while Letitia delves into the limitations of deep learning models in tasks like visual question answering due to biases in data sets. The discussion highlights the potential of multimodality to address the unrobustness of deep learning models.In this clip
From this podcast

Machine Learning Street Talk (MLST)
#70 - LETITIA PARCALABESCU - Symbolics, Linguistics [UNPLUGGED]
Related Questions