A linear model tree is a decision tree with a linear functional model in each sheet, while in a classical regression tree (e.g. CART) this is the average sample value for the response variable for statistical units in each sheet (therefore, a constant), which is considered. Linear tree models can be considered as a form of locally weighted regression, and a regression tree is a piecewise constant regression.
For more information on linear model trees, you can refer to
Torgo, L. . 14- , . 385-393. , 1997 .