AI Paper Atlas

Research map / Continual learning / Class-incremental

DiffClass: Diffusion-Based Class Incremental Learning

Zichong Meng, Jie Zhang, Changdi Yang, Zheng Zhan, Pu Zhao, Yanzhi Wang

ECCV 2024 · Continual learning › Class-incremental · approach: Exemplar-efficient class-incremental learning · 27 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
replace baseline synthetic-data CIL with DiffClassexemplar-free CIL on CIFAR100 · CIFAR100 N=20Acc_L: 42.87 -> 57.11 vs SEEDImproved
replace SEED with DiffClassexemplar-free CIL on ImageNet100 · ImageNet100 N=20Acc_L: 45.77 -> 68.68Improved
add multi-distribution matching (MDM) diffusion fine-tuning (ablation)DiffClass · ImageNet100 N=5Acc_avg: 59.71 -> 74.85Improved
add multi-domain adaptation (domain classifier with gradient reversal) (ablation)DiffClass · ImageNet100 N=5Acc_avg: 65.29 -> 74.85Improved
add selective synthetic image augmentation (SSIA) (ablation)DiffClass · ImageNet100 N=5Acc_avg: 62.37 -> 74.85Improved
train with synthetic previous-task data instead of real dataResNet with linear classifier · ImageNet100 20-task diagnosis, first four taskstotal accuracy: 82.72 -> 59.37Hurt
increase number of tasks N from 10 to 20SEED · CIFAR100Acc_L: 51.42 -> 42.87Hurt
increase number of tasks N from 10 to 20DiffClass · CIFAR100Acc_L: 58.40 -> 57.11No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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