Inclusion of DPP loss as summary length regularization doesn't help in quality summarization

[Anurag14]

The authors propose DPP loss for Diversity Regularization in their model. Detrimental Point processes are a idea that help in sampling diverse subset of points from a set of points. This is a dire extension without which the model implementation is incomplete. I am willing to help in this. So can you raise a ticket about things to do and add it in read me. Cheers

[idlerm]

Where is the 360 dataset downloaded?

[idlerm]

Where is the 360 dataset downloaded? Thank you very much for your code.

[mpalaourg]

@Anurag14 A bit late to the party, but... do you have a working implementation of the dpp loss? I cant get over the paper statement for the L(s)

L(s) is a smaller square matrix, cut down from L given s.

s i assume is the generated summary of 0 and 1, but how can i pick a small square matrix? Even worse, what to do if s is not binary but real valued in (0, 1)?