Search behaves strangely when scoring

hi,

could someone please explain why the document shows up as a duplicate, yet the other (duplicated) document doesn’t show up in the search?

And how is it possible that the website [1] is higher rated than the actual PDF?

thank you.

[1] research.microsoft.com/apps/pubs … ?id=231106

i think i just answered #1: is it possible that during a search, the title heavily influences the score? if i rename the duplicate to “a” it is not even in the top 50 search results, the other one is #7.