Controllable Data Generation with Hierarchical Neural Representations

Sheyang Tang, Xiaoyu Xu, Jiayan Qiu, Zhou Wang

Keywords: Implicit Neural Representations, Generative INR, Generative Models, Diffusion Models

Abstract: Implicit Neural Representations (INRs) represent data as continuous functions using the parameters of a neural network, where data information is encoded in the parameter space. Therefore, modeling the distribution of such parameters is crucial for building generalizable INRs. Existing approaches learn a joint distribution of these parameters via a latent vector to generate new data, but such a flat latent often fails to capture the inherent hierarchical structure of the parameter space, leading to entangled data semantics and limited control over the generation process. Here, we propose a $\textbf{C}$ontrollable $\textbf{H}$ierarchical $\textbf{I}$mplicit $\textbf{N}$eural $\textbf{R}$epresentation ($\textbf{CHINR}$) framework, which explicitly models conditional dependencies across layers in the parameter space. Our method consists of two stages: In Stage-1, we construct a Layers-of-Experts (LoE) network, where each layer modulates distinct semantics through a unique latent vector, enabling disentangled and expressive representations. In Stage-2, we introduce a Hierarchical Controllable Diffusion Model (HCDM) to capture conditional dependencies across layers, allowing for controllable and hierarchical data generation at various semantic granularities. Extensive experiments on CelebA-HQ, ShapeNet, SRN-Cars, and AMASS datasets demonstrate that CHINR improves generalizability and offers flexible hierarchical control over the generated content.

Primary Area: generative models

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 5442