AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Joint Enhancement and Classification using Coupled Diffusion Models of Signals and Logits

Gilad Nurko, Roi Benita, Yehoshua Dissen, Tomohiro Nakatani, Marc Delcroix, Shoko Araki, Joseph Keshet

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data · 0 citations (OpenAlex)

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
couple logit diffusion with signal diffusion (Parallel) instead of enhance-then-classifyParallel · CIFAR-100 30% GNaccuracy: 60.4 -> 68.5Improved
couple logit diffusion with signal diffusion (Nested)Nested · CIFAR-10 30% GNaccuracy: 60.4 -> 81.2Improved
couple logit diffusion with signal diffusion (Parallel)Parallel · IN32-100 30% GNaccuracy: 50.6 -> 65.5Improved
couple diffusion on MNIST instead of URIENested · MNIST 30% GNaccuracy: 99.3 -> 99.1Hurt
couple logit diffusion with signal diffusion (Nested) for ASRNested · EARS-WHAMWER (%): 10.04 -> 9.75Improved
couple logit diffusion with signal diffusion (Nested) for ASRNested · EARS-ReverbWER (%): 4.48 -> 3.83Improved
use Dissen enhancer trained with task loss instead of audio-only enhancementDissen · EARS-ReverbWER (%): 4.48 -> 6.46Hurt
condition on clean estimate instead of noisy sample in Parallel (ablation)Parallel · CIFAR-100 30% GNaccuracy: 67.0 -> 68.5Improved
condition on clean estimate instead of noisy sample in Parallel (ablation)Parallel · IN32-100 Gaussian Bluraccuracy: 61.8 -> 63.3Improved
use DDPM stochastic sampling instead of DDIM in Parallel (ablation)Parallel · CIFAR-100 Gaussian Bluraccuracy: 64.8 -> 69.7Improved
use DDPM stochastic sampling instead of DDIM in Parallel (ablation)Parallel · CIFAR-100 15% GNaccuracy: 70.7 -> 73.5Improved
scale to 224x224 ImageNet 10-class with ParallelParallel · ImageNet 224 30% GNaccuracy: 86.5 -> 92.4Improved

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