AI Paper Atlas

Research map / Continual learning / Class-incremental

Random Amalgamation of Adapters for Flatter Loss Landscapes: Towards Class-Incremental Learning with Better Stability

Yao Deng, Xiang Xiang, Jiaqi Gui

AAAI 2026 · Continual learning › Class-incremental · approach: Exemplar-efficient class-incremental learning · 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
add randomized adapter amalgamation (RAA) (ablation)adapter CIL baseline · ImageNetAlast accuracy: 62.43 -> 64.01Improved
add randomized adapter amalgamation (RAA) (ablation)adapter CIL baseline · ImageNetRlast accuracy: 79.38 -> 80.22Improved
replace SAM with SAM+ (ablation)adapter CIL baseline · ImageNetAlast accuracy: 63.53 -> 63.86Improved
add vanilla SAM (ablation)adapter CIL baseline · ImageNetAlast accuracy: 62.43 -> 63.53Improved
add SAM+ on top of RAA (ablation)RAA · ImageNetRlast accuracy: 80.22 -> 80.31Improved
add SAM+ plug-in to existing CIL methodsEase, RanPAC, SLCA, SSIAT and others · mean of 4 datasetslast accuracy: Ease 75.88 -> 76.35Improved
increase adapters from 2 to 4 (ablation)RAA · ImageNetRlast accuracy: 80.31 -> 78.05Hurt
increase adapters from 2 to 3 (ablation)RAA · CUB200last accuracy: 89.94 -> 90.18No clear effect
RAA vs strongest baselinesRAA · five CIL benchmarksaverage of last accuracy: 78.24 -> 79.73Improved
ViT-B/16-1K instead of 21K backbone (ablation)adapter CIL · ImageNetAlast accuracy: 62.43 -> 64.88 (ImageNetA); 91.48 -> 90.72 (CIFAR100)Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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