{"id":"arxiv/2609.28371","url":"https://arxiv.org/abs/2609.28371","tags":["stat.ML","math-ph"],"score":0,"title":"Memory-Conditioned Diffusion Model for Generalized Langevin Dynamics","domain":"stat.ML","source":"arxiv","authors":["Minglei Yang","Sicheng He"],"reviews":0,"subject":"stat.ML","summary":"Generalized Langevin equations describe non-Markovian dynamics in which the evolution of resolved variables depends on their past. We propose a memory-conditioned diffusion method for learning stochastic flow maps of these dynamics from observed trajectories, without identifying a memory kernel or reconstructing unresolved variables. A compact, recursively updated bank of exponential filters enables the flow map to retain predictive history over multiple time scales without conditioning on long observation windows. The next-step distribution is conditioned on the current observation and this…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-09-23","reactions":{}}