The paper investigates a class of partial differential equations involving a nonlocal operator called the fractional Laplacian, which generalizes ordinary diffusion by allowing long-range interactions. Specifically, the authors study equations of the form where this operator acting on an unknown function equals a nonlinear term, all set inside a bounded, nicely shaped region with the function forced to be zero outside it. The central question is whether all positive solutions to such equations are automatically bounded by a universal constant that does not depend on the particular solution chosen. These kinds of bounds, called uniform a priori estimates, are important because they are a key step toward proving that solutions actually exist and understanding their behavior.
The difficulty comes from the specific form of the nonlinearity, which sits right at the so-called critical exponent. This is a threshold value that separates well-behaved problems from problematic ones. Near this threshold, solutions can develop highly concentrated spikes, a phenomenon called bubbling, where mass piles up at a point and a uniform bound seems ready to break down. This makes the slightly subcritical case significantly harder than the strictly subcritical one, where uniform bounds have been known for some time.
The main contribution of the paper is identifying a precise condition on a slowly varying perturbation of the nonlinearity that prevents this bubbling from occurring. Intuitively, the condition requires the perturbation to grow or decay in a way that acts as a barrier against mass concentration as the solution gets large. When this condition holds, the authors prove that all positive solutions remain uniformly bounded, regardless of which solution you pick. This considerably widens the catalog of equations for which such estimates are rigorously established and provides a clearer picture of exactly what structural features of the nonlinearity control the regularity of solutions near the critical threshold.