A theoretical basis for model collapse in recursive training

Vivek S. Borkar

doi:10.48550/arxiv.2506.09401

June 11, 2025Open Access

A theoretical basis for model collapse in recursive training

View Full Paper

Ask AI

Bookmark

Share

Key Points

Model collapse occurs in recursive training, producing distinct outcomes based on sample contributions.
Two different asymptotic behaviours are identified associated with the presence of any external source.
Understanding these behaviours can help refine recursive training methods to mitigate collapse.
This theoretical basis provides insights for future research in generative models and their stability.

Implication

Theoretical exploration reveals contrasting asymptotic behaviours of probability distribution with external sources.

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.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Vivek S. Borkar (2025) studied this question.

synapsesocial.com/papers/68f6196ee0bbbc94fac3621f https://doi.org/10.48550/arxiv.2506.09401

View Full Paper

Ask AI

Bookmark

Share

A theoretical basis for model collapse in recursive training

Key Points

Implication

Abstract

Discussion

Cite This Study

A theoretical basis for model collapse in recursive training

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

Keep reading

More related papers