Jing Yu Koh

@gberta When I run the demo BNF_multi_class_edge_affinity_demo.m for the train image, (2008_000003.png), the output appears slightly different from the paper. Is this due to implementation differences in the code published?...

They're mostly independent, you can refer to Table 3 in the appendix of the paper for an ablation. We find that the captioning loss doesn't really affect retrieval performance that...

Hi, thanks for pointing this out! I realized this wasn't mentioned in the paper (we'll add it to the next arXiv version), but we do the same as [MAGMA](https://arxiv.org/abs/2112.05253) and...

It should be the longest answer for each question.

Glad you solved it. Yeah, I think the right way to do this would be to add this as a param to the model args.