{"id":"arxiv/2605.17082","url":"https://arxiv.org/abs/2605.17082","tags":["math.PR"],"score":0,"title":"Spectral Purification in Reversible Markov Chains: Hidden Parameters, Observable Equivalence, and Finite-Time Rigidity","domain":"math.PR","source":"arxiv","authors":["Qiao Wang"],"reviews":0,"subject":"math.PR","summary":"Classical spectral theory of reversible Markov chains characterizes the asymptotic relaxation dominated by the slowest eigenmode, governed by the spectral gap. This paper studies a complementary finite-time phenomenon: long before stationarity, many observables of the relaxation trajectory behave as if only a single spectral mode remains -- a collapse we term spectral purification. We develop a quantitative theory for it. The theory rests on three structural insights: (i) shifting focus from the state trajectory to the modal distribution $p_i(k)$ (normalized spectral energy across modes);…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-16","reactions":{}}