← Search

Marta Aparicio Rodriguez

2 accepted papers

2026

Diffusion Models Preferentially Memorize Prototypical Examples or: Why Does My Diffusion Model Love Slop?

ICML 2026poster

Generative models have a persistent limitation: their tendency to memorize training data can create legal liabilities and erode creative diversity. Understanding which samples are memorized in whole or in part, and under what conditions, therefore remains an important open problem. Here we answer th…

Cited by 0SourceScholar
2025

Concept Reachability in Diffusion Models: Beyond Dataset Constraints

ICML 2025poster

Despite significant advances in quality and complexity of the generations in text-to-image models, *prompting* does not always lead to the desired outputs. Controlling model behaviour by directly *steering* intermediate model activations has emerged as a viable alternative allowing to *reach* concep…