On the Interpretability of Attention Networks

1 code implementation • 30 Dec 2022 • Lakshmi Narayan Pandey, Rahul Vashisht, Harish G. Ramaswamy

In trained models with an attention mechanism, the outputs of an intermediate module that encodes the segment of input responsible for the output is often used as a way to peek into the `reasoning` of the network.

Image Captioning