Ricardo U.

Happy to redirect to one of my PhD students or colleagues from HITeC e.V. if you are still looking for someone

Indeed, we are also looking into automatic prompt tuning to make it fit - there are some tricks such as vocabulary reducing etc but we noticed that longer, complex queries are kind of hard

well, I think both perform the same but chatGPT might be a little bit better

In general, it helps to give 30 or more examples as a prefix

Yes, in general, the LLM quality of SPARQL generation is really bad, no matter what you do or which LLM you use.

Hi, sorry for the self-plugging: https://arxiv.org/abs/2204.12793 and https://ceur-ws.org/Vol-3592/ should give you a zoo of possible models

Hi, for Germany, I would support that endeavour. I guess, we can even find a large number of KG-enthusiasts to make that happen? Harald Sack Sebastian Hellmann …?

I do not see the closing brackets, could be the source of error. My guess is that ?x is always bound in the first UNION and thus you would need to reformulate to use OPTIONAL instead of UNION and move ?x rdf:type ?restriction outside

Hi, sure, that should work. Do you get an error?