... C 1 C 2= C −1 C 1 C −1 C 1=Lexical-target Instances C 0 C n(n = −2 2) C 0 C −2=, C 0 C −1=, C 0 C 0=, C 0 C 1=, C 0 C 2= C 0 C n C n+1(n = −2 1) C 0 C −2 C −1=, C 0 C −1 C 0=, ... has comparable performance898Non-lexical-target Instances C n(n = −2 2) C −2=, C −1=, C 0=, C 1=, C 2= C n C n+1(n = −2 1) C −2 C −1=, C −1 C 0=, C 0 C 1=, C 1 C 2= C −1 C 1 C −1 C 1=Lexical-target ... through the same approach.To facilitate tuning the weights, we use two com-ponents of the co-occurrence model Co(W, T ) torepresent the co-occurrence probability of W and T ,rather than use Co(W,...