{"id":"arxiv/2607.08757","url":"https://arxiv.org/abs/2607.08757","tags":["stat.ML","cs.LG","math.NA","math.PR"],"score":0,"title":"Score Accuracy Along the Forward Diffusion Does Not Certify Numerical Stability in Diffusion Sampling","domain":"stat.ML","source":"arxiv","authors":["Yiwei Zhou"],"reviews":0,"subject":"stat.ML","summary":"Score matching controls average error under the forward marginals, but a discretized reverse-time sampler evaluates the learned score along its own trajectory. We show that small forward-marginal error does not guarantee numerical stability. We construct a single smooth score field with arbitrarily small forward-marginal $L^2$ error. The learned reverse-time process is nonexplosive, has moments of every order, and can be arbitrarily close to the exact reverse-time process in path-space total variation. Yet its Euler--Maruyama discretizations converge in probability while every positive…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-07-09","reactions":{}}