A Conditional Random Field $($CRF$)$ model for Optical Character Recognition $($OCR$)$ with only singleton and pairwise potentials is an instance of a template model, with variables C$1,...,$Cn$.$over the characters and observed images I$1,...,$In$.$ The model used is a template model in that the singleton potentials are replicated across different Ci variables, and the pairwise potentials are replicated across character pairs. The structure of the model is shown below:

Now consider the advantages of this particular template model for the OCR task, as compared to a non$-$template model that has the same structure, but where there are distinct singleton potentials for each Ci variable, and distinct potentials for each pair of characters. Which of the following about the advantage of using a template model is true? $1$ or more options may be correct.

$>$The same template model can be used for words of different lengths.

$>$Parameter sharing could make the model less susceptible to over$-$fitting when there is less training data.

$>$The inference is significantly faster with the template model.

$>$The template model can incorporate position$-$specific features, e$.$g$.$ q$-$u occurs more frequently at the beginning of a word, while a non$-$template model cannot.