deletions | additions       diff --git a/To_maintain_a_high.tex b/To_maintain_a_high.tex  deleted file mode 100644  index e51497e..0000000  --- a/To_maintain_a_high.tex  +++ /dev/null 
...
To maintain a high precision level of answers, the Syphon system should be  conservative about the questions it allows. For the baseline system,  in light of the analysis of question classes above,  we want to propose a simple Syphon which scans the question  for named entities%  \footnote{\textit{Named entity} in NLP parlance is something that is  not a plain English word carrying meaning, but e.g.\ a proper  name or a numerical value (date, monetary amount, \dots).  There exist specialized scanners that extract named entities  of specific types; as a simple baseline, we may consider  all strings that are titles of Wikipedia articles (or redirects,  that is essentially unique aliases). For example  \textit{Hillary Clinton} and \textit{President of the United States}  both are titles of such articles.}  and verbs which describe clear relationships  and verifies that no words without an assigned role remain  in the question. This should keep a fair share of the sample questions  covered and with a clear interpretation.  Fig.~\ref{fig:syphon} shows a simple visual concept of the initial Syphon.  diff --git a/layout.md b/layout.md  index 1b65108..7cd9ef5 100644  --- a/layout.md  +++ b/layout.md 
...
sectionStructure_of_.tex  beginitemize__item_C.tex  sectionHow_Can_We_Kn.tex  To_maintain_a_high.tex  sectionA_Concrete_Pr.tex  beginitemize__item_S.tex  sectionImplementatio.tex