Radial specific star formation rate (sSFR) gradients are crucial for determining how stellar mass is built up in galaxies as they evolve along the star formation main sequence (SFMS), and how and where the star formation shuts down as galaxies move off the SFMS to become fully quiescent. However, dust gradients have often been overlooked or treated to be subordinate when studying distant galaxies, which could be seriously distorting our view of the structure of distant galaxies without proper dust corrections. Although spectroscopic approaches perform better in disentangling the age(sSFR)-dust-metallicity degeneracy, they are largely limited by telescope time, spectrum quality and sample size, especially for distant galaxies. Now thank to the rich HST deep multi-band photometry data from HST/CANDELS, the age(sSFR)-dust-metallicity degeneracy can be partially broken up by fitting attenuated stellar population models to broad-band spectral energy distributions or by analyzing two-color (i.e., UVJ, UVI) diagrams of galaxies. In my talk, I will present a brief review of a series of works about the radial sSFR profiles of distant galaxies by our group, which are determined using the new multi-band multi-aperture photometry catalogs of the CANDELS. Our studies give a new and powerful insight into how distant galaxies assemble as they evolve along the SFMS, and how the star formation shuts down as they move off the SFMS. Finally, I will discuss the physical processes responsible for cessation of star formation in distant galaxies.