AI Paper Atlas

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

COLA: Continual Orthogonal Low-Rank Adaptation for Class-Incremental Learning

Monu Nagar, Debasis Das

ECCV 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 soft orthogonal projection matrix P to LoRA (ablation)COLA · ImageNet-A, ImageNet-R, CUB200Acc: +2.6% on ImageNet-A, 2.7% on ImageNet-R, 3.2% on CUB200Improved
apply LoRA to query/value instead of key/value (ablation)COLA · ImageNet-R (10 tasks)Acc (%): 77.13 -> 77.56Improved
apply LoRA to query/value instead of key/value (ablation)COLA · ImageNet-A (10 tasks)Acc (%): 57.86 -> 59.20Improved
apply LoRA to query/value instead of key/value (ablation)COLA · CUB200 (20 tasks)Acc (%): 68.17 -> 69.18Improved
set residual scaling gamma to 0.20 instead of 0.60 (ablation)COLA · ImageNet-R (10 tasks)Acc (%): 73.23 -> 77.56Improved
set residual scaling gamma to 0.20 instead of 0.08 (ablation)COLA · ImageNet-A (10 tasks)Acc (%): 56.55 -> 59.20Improved
use ViT-B/16 instead of DeiT-S/16 backboneCOLA · ImageNet-R (10 tasks)Acc (%): 53.92 -> 77.56Improved
use ViT-B/16 instead of DeiT-B/16 backboneCOLA · CUB200 (10 tasks)Acc (%): 65.22 -> 74.51Improved
use COLA instead of SD-LoRAImageNet-R (25 tasks) · ImageNet-R (25 tasks)AAA (%): 74.71 -> 82.11Improved
use COLA instead of VPT-NSP2ImageNet-R (10 tasks) · ImageNet-R (10 tasks)Acc (%): 78.25 -> 77.56Hurt
use COLA instead of InfLoRA/SD-LoRA for computeCOLA · EfficiencyGFLOPs: 35.12 -> 32.22Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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