AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Diffusion and Flow-based Copulas: Forgetting and Remembering Dependencies

David Huk, Theodoros Damoulas

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and learned samplers

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use classification-diffusion copula instead of Ratio copula for densitycopula density estimation · Magiclog-likelihood: 6.76 -> 18.65Improved
use classification-diffusion copula instead of Ratio copulacopula density estimation · Dry_Beanlog-likelihood: 48.21 -> 50.21Improved
use classification-diffusion copula instead of Ratio copulacopula density estimation · MNISTlog-likelihood: 334.42 -> 346.70Improved
use reflection or c_dc copula instead of Gaussian/vine for samplescopula sampling · RobocupWasserstein-2: 3.96 -> 3.84Improved
use proposed copulas instead of Gaussian copula for Kendall-tau fitdependence structure · RobocupFrobenius norm: 0.45 -> 0.49 (reflection), 0.51 (c_dc)Hurt
use c_dc instead of Gaussian copulasample quality · MNISTFID: 102.56 -> 7.38Improved
use reflection copula instead of best baselinesample quality · Cifar cars/trucksFID: 100.04 -> 42.14Improved
use c_dc instead of Gaussian copula on digits FIDsample quality · digitsFID: 5.74 -> 15.24Hurt
use reflection copula instead of Gaussian on Cifar W2sample quality · Cifar cars/trucksWasserstein-2: 30.62 -> 32.40Hurt
use vine copula for densityCifar density · Cifar cars/truckslog-likelihood: NaNHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy