Pith. sign in

REVIEW 5 cited by

DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2301.13721 v3 pith:E75VHD3A submitted 2023-01-31 cs.CV

Tao Yang , Yuwang Wang , Yan Lv , Nanning Zheng This is my paper
classification cs.CV
keywords dpmsfactorsdiffusiondisdiffdisentangledfieldsrepresentationtask
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Targeting to understand the underlying explainable factors behind observations and modeling the conditional generation process on these factors, we connect disentangled representation learning to Diffusion Probabilistic Models (DPMs) to take advantage of the remarkable modeling ability of DPMs. We propose a new task, disentanglement of (DPMs): given a pre-trained DPM, without any annotations of the factors, the task is to automatically discover the inherent factors behind the observations and disentangle the gradient fields of DPM into sub-gradient fields, each conditioned on the representation of each discovered factor. With disentangled DPMs, those inherent factors can be automatically discovered, explicitly represented, and clearly injected into the diffusion process via the sub-gradient fields. To tackle this task, we devise an unsupervised approach named DisDiff, achieving disentangled representation learning in the framework of DPMs. Extensive experiments on synthetic and real-world datasets demonstrate the effectiveness of DisDiff.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Steering Optimisation Trajectories in Diffusion Representation Learning

    cs.CV 2026-07 conditional novelty 6.0 of 10

    SteeringDRL identifies two optimization regimes in diffusion autoencoders and uses gated residual U-Nets with a log SNR curriculum to steer training toward disentangled representations, improving performance across mu...

  2. Harnessing Text-to-Image Diffusion Models for Point Cloud Self-Supervised Learning

    cs.CV 2025-07 conditional novelty 6.0 of 10

    PointSD uses a frozen Stable Diffusion model, conditioned on point clouds through rendered images, to generate training targets for point cloud self-supervised learning.

  3. Diffusion Counterfactual Generation with Semantic Abduction

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Diffusion-based causal image counterfactuals with semantic abduction improve identity preservation at a small cost in intervention effectiveness, demonstrated on Morpho-MNIST, CelebA-HQ, and mammogram artifact removal.

  4. Decafs: Disentangled Conditional adversarial Flows

    cs.LG 2026-07 conditional novelty 5.0 of 10

    Decaf trains a Lie-group generator whose disentangled latent space is aligned with a normalizing flow's latent space, so that changing one learned coordinate changes one generative factor.

  5. Riemannian Deep Learning: Modules, Networks, and Geometries

    cs.LG 2026-07 accept novelty 4.0 of 10

    One Lie-group/gyrogroup framework unifies batch normalization and logistic-regression classifiers across SPD, rotation, correlation, Grassmannian, and constant-curvature manifolds, with new hyperbolic and SPD geometries.

Pith tools