Liouville Quantum Gravity (LQG) is a mathematical framework for constructing random surfaces that appear in string theory and the study of two-dimensional quantum gravity. The basic idea is to take the ordinary flat plane and randomly warp distances using a rough random function called a Gaussian free field, controlled by a parameter gamma. When gamma is strictly between 0 and 2, this warping produces what is called the subcritical LQG metric, a well-understood random way of measuring distances on the plane. The case gamma equals 2 is called critical and is harder to analyze because the randomness becomes much more extreme right at that boundary.
The central result of this paper is that as gamma approaches 2 from below, the subcritical LQG metrics converge to the critical 2-LQG metric. Because the natural scale of these metrics blows up near the critical point, one has to normalize them by appropriate rescaling factors before taking the limit. Once that is done, the metrics converge in a strong sense: the distance functions, viewed as continuous functions of pairs of points in the plane, converge in distribution to the critical metric. The convergence holds in the local uniform topology, meaning that on any bounded region the random distance functions become arbitrarily close to the critical one.
Combined with a companion result showing convergence as gamma decreases to 0 (where the LQG metric reduces to ordinary Euclidean distance), this paper completes a picture of full continuity across the entire range of gamma from 0 to 2. In other words, the family of LQG metrics forms a continuous one-parameter family connecting flat Euclidean geometry at one end to critical quantum gravity geometry at the other. This kind of continuity result is important for understanding how the geometry of random surfaces changes as the underlying randomness is tuned, and it provides rigorous mathematical foundations for intuitions that physicists have long held about these models.