about the abstract

[arademaker]

I will repeat here a question that I sent to @oepen by email:

You wrote:

Despite remarkable advances in vector-based (continuous, dense, and distributed) encodings of meaning, ‘classic’ (hierarchically structured and discrete) semantic representations will continue to play an important role in ‘making sense’ of natural language.

Do you have references for this claim? I really agree on that, but many people keep asking for pieces of evidence to support the claim that discrete representations will continue to play an important role...

[oepen]

inasmuch as that quote in fact is a ‘claim’, it is a view into the crystal ball of the authors, i.e. our expectation about future developments. thus, most of the supporting references have yet to be written :-). but the final part of the tutorial briefly reviews some applications where semantic structure has been put to good use.

i am leaving this issue open for now, in the hope that others might chime in.