{"id":"arxiv/2605.13448","url":"https://arxiv.org/abs/2605.13448","tags":["stat.ML","cs.LG","math.PR"],"score":0,"title":"On the Limits of Latent Reuse in Diffusion Models","domain":"stat.ML","source":"arxiv","authors":["Yifeng Yu","Lu Yu"],"reviews":0,"subject":"stat.ML","summary":"Diffusion models are often trained in low-dimensional latent spaces, which are then reused for related but shifted datasets. In this work, we study when such latent reuse remains reliable under distribution shift. We consider a source-target setting in which both datasets are approximately low-dimensional but may lie near different subspaces. We show that freezing and reusing a source latent space induces a target-domain score error governed by two quantities: the principal-angle misalignment between the source and target subspaces, and the target ambient noise amplified by the diffusion…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-13","reactions":{}}