Recently, several moderate to large earthquakes challenged our understanding of earthquake mechanics, and the recent availability of large geophysical datasets revealed that what we commonly call earthquakes manifest a complexity that is far beyond our comprehension. This session focuses on experimental, numerical, and field studies providing new details on fault slip acceleration from all perspectives and scales. We encourage the presentation of works encompassing novel methods or the combination of different techniques, such as shear-stress or energy controlled experiments, the combination of deformation experiments and imaging techniques, machine learning applied to experimental and field data. The session should stimulate discussion among communities that study earthquakes and friction in geomaterials. We envision the opportunity for such communities to benefit from one another’s contributions to expand the capability of upscaling observations from the laboratory and numerical simulations, improving our understanding of the spectrum of fault failure behaviors.