2000 character limit reached
A theoretical basis for model collapse in recursive training (2506.09401v2)
Published 11 Jun 2025 in math.PR and cs.LG
Abstract: It is known that recursive training from generative models can lead to the so called `collapse' of the simulated probability distribution. This note shows that one in fact gets two different asymptotic behaviours depending on whether an external source, howsoever minor, is also contributing samples.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.