At this moment, I remember this recent paper that use Google Earth images for analyzing gully erosion in Greece -> Article Towards an Assessment of the Ephemeral Gully Erosion Potenti...
I do not know right now if there are paper related to landslides and Google Earth images.
My student Patrisse Vasek and I used GoogleEarth to prospect for lineaments (potential faults and joints serving as contaminant pathways). Check my publication profile for tha abstract. My team of expert witnesses have been using GoogleEarth in ongoing interventions against uranium mining, monitoring ISL uranium mining leases for unpermitted durface dumping of spent lixiviant. Check the several opinions and hearing transcripts on my profile. I'm still using GoogleEarth imagery to do this work in Sioux, Dawes, Box Butte, and Sheridan counties in Nebraska, and Fall River, Custer, Oglala Lakota, Pennington, and Harding counties in South Dakota.
I'd be careful when using Google Earth to analyze temporarily varying things like landslides. The Google Earth cloud free image is stitched together from different scenes at different times. While it should be no problem mapping slowly evolving features such as faults, for landslide mapping this could become an issue.
There are several papers that I found when searching the following within the search engine = (landslide mapping + google earth images + google scholar). Following is an example that I found. Hope this helps.
(An integrated approach to map landslides in Chittagong Hilly Areas, Bangladesh, using Google Earth and field mapping)