{"id":"arxiv/2604.20514","url":"https://arxiv.org/abs/2604.20514","tags":["math.PR","math-ph"],"score":0,"title":"Macroscopic loops in the random loop model on sparse random graphs","domain":"math.PR","source":"arxiv","authors":["Andreas Klippel"],"reviews":0,"subject":"math.PR","summary":"We study the random loop model with crosses and bars on sparse random graphs. Our main objective is to prove the existence of macroscopic loops, in the sense that a loop visits a positive proportion of the vertices. We develop a deterministic drift method on arbitrary finite graphs based on three ingredients: a local split--merge--rewire analysis for the loop number, an exact differential identity for the partition function, and a slice estimate reducing the relevant same-loop insertion volume to induced edge counts of small vertex sets. This yields a general criterion in terms of a…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-04-22","reactions":{}}