Image Text Retrieval Loss Discussion

Hi.

Why did you decide to use a BCE loss on the ITM pretraining text and a ranking loss on the ITM downstream task? Is there any intuition behind this? Why not use a ranking loss on both?

Thanks for the great work.

-Caleb

No file, test, or entry point is named. Start by locating the image-text matching pretraining and downstream ranking-loss implementations, then read the corresponding sections of the UNITER paper. Done means documenting the rationale for the different losses and explaining why a ranking loss is not used in both settings.

Written by the indexing model from the issue text.