The thought arises that Prolog already has continuations but they are not first-class objects:
?- member(X,[1,2,3]). X = 1 ; % <-- your continuation is waiting X = 2 ; % <-- you called it X = 3. % <-- you called it and it's done
The paper "Schrijvers et al., 2013" has been published in Theory and Practice of Logic Programming in 2013.
I like this one more:
"Delimited continuations in Prolog: semantics, use, and implementation in the WAM"
- Tom Schrijvers (UGent), Bart Demoen and Benoit Desouter (UGent)
- March 13, 2013
Technical Report. This is an excellent resource.
A lot of notes on Delimited Continuations
Right here, including example code.
(But I didn't go into the problems with the cut described in the text.)