AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

LCA: Local Classifier Alignment for Continual Learning

Tung Tran, Danilo Vasconcellos Vargas, Khoat Than

ICLR 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation · 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 Local Classifier Alignment (LCA) after incremental merging (ablation)IM (ViT-B/16-IN1K) · Overall (7 datasets)average accuracy: 80.9 -> 85.6Improved
add LCA after incremental merging (ablation)IM · ImageNet-Aaverage accuracy: 66.5 -> 75.0Improved
add LCA after incremental merging (ablation)IM · VTABaverage accuracy: 84.6 -> 95.2Improved
add LCA after incremental merging (ablation)IM · OmniBenchmarkaverage accuracy: 81.1 -> 81.4No clear effect
use LCA robustness term instead of plain classifier alignment (CA) (ablation)IM Max · ImageNet-Aaverage accuracy: 70.86 -> 73.73Improved
use LCA instead of CA (ablation)IM MaxAbs · VTABaverage accuracy: 91.71 -> 95.2Improved
add classifier alignment (CA) after merging (ablation)IM Max · ImageNet-Aaverage accuracy: 66.26 -> 70.86Improved
choose Min, Max or MaxAbs merge operator (ablation)IM+LCA · ImageNet-A / VTAB / CARSaverage accuracy: 73.00-75.0 (IN-A)No clear effect
add LCA to MOS and SLCA as final alignment step (ablation)MOS, SLCA · seven CIL datasetsaverage accuracyImproved
add LCA to IM (ablation)IM · CIFAR100-C and CIFAR100-Probustness accuracy: +2% (C), +2.5% (P)Improved
make robustness penalty lambda too strong (ablation)LCA · CIFAR100 and another datasetaccuracyHurt
add LCA to VPT-based PEFT merging (ablation)IM with VPT · VTABaccuracyImproved

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