Detection of Memorization in Diffusion Models Through Thermodynamic Singularity
En palabras de los autores
In diffusion models, proper distributional learning entails reversing a diffusion process over underlying density functions, whereas a key failure is memorization, in which the reverse process instead learns diffusion from individual training samples. We provide an analytical characterization of such memorized reverse processes, along with a criterion that detects the associated thermodynamic singularity. The proposed criterion reliably distinguishes between properly learned and memorizing networks across a range of neural network models designed to promote or mitigate memorization.
Resultado principalEl resumen no menciona limitaciones.
Apareció: miércoles, 21 de octubre. arXiv. Preprint, todavía sin revisión por pares.