Yu-Han Wu — accepted AI-conference papers

2026

Optimal Stopping in Latent Diffusion Models

Yu-Han Wu, Quentin Berthet, Gérard Biau, Claire Boyer, Romuald Elie, Pierre Marion

ICML 2026poster

We identify and analyze a surprising phenomenon of $\textit{Latent}$ Diffusion Models (LDMs) where the final steps of the diffusion can $\textit{degrade}$ sample quality. In contrast to conventional arguments that justify early stopping for numerical stability, this phenomenon is intrinsic to the di…

Cited by 0 Source ↗Scholar ↗

2024

Implicit regularization of deep residual networks towards neural ODEs

Pierre Marion, Yu-Han Wu, Michael Eli Sander, Gérard Biau

ICLR 2024spotlight

Residual neural networks are state-of-the-art deep learning models. Their continuous-depth analog, neural ordinary differential equations (ODEs), are also widely used. Despite their success, the link between the discrete and continuous models still lacks a solid mathematical foundation. In this arti…

Cited by 20 Source ↗PDF ↗code ↗Scholar ↗