Local Linear Forests (2018, methodology): “…helpful to use a local regression adjustment to correct for potential misalignment between a test point and its neighborhood (Stone 1977; Tibshirani and Hastie 1987; Cleveland and Devlin 1988; Newey 1994; Fan and Gijbels 1996; Heckman, Ichimura, and Todd 1998; Loader 1999; Abadie and Imbens 2011).”
Interference Between Units in Randomized Experiments (2007, methodology): “The curve in the upper left panel in Figure 1 is the lowess smooth of responses on xi ignoring treatment group (Cleveland and Devlin 1988, as implemented in R), and the plot in the upper right panel displays the residuals from the smooth.”