AI Paper Atlas

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

Incremental Task Learning with Incremental Rank Updates

Rakib Hyder, Ken Shao, Boyu Hou, Panos Markopoulos, Ashley Prater-Bennette, M. Salman Asif

ECCV 2022 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation · 17 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
raise initial rank from 1 to 6, keeping increment 1 (ablation)low-rank ITL · P-MNISTaverage accuracy (%): 74.23 -> 82.21Improved
raise initial rank from 6 to 11, keeping increment 1 (ablation)low-rank ITL · P-MNISTaverage accuracy (%): 82.21 -> 85.61Improved
raise initial rank from 1 to 11, keeping increment 1 (ablation)low-rank ITL · R-MNISTaverage accuracy (%): 81.57 -> 91.09Improved
raise rank increment per task from 1 to 2 (initial rank 11) (ablation)low-rank ITL · P-MNISTaverage accuracy (%): 85.61 -> 90.51Improved
raise rank increment per task from 2 to 4 (initial rank 11) (ablation)low-rank ITL · P-MNISTaverage accuracy (%): 90.51 -> 93.84Improved
raise rank increment per task from 1 to 4 (initial rank 11) (ablation)low-rank ITL · R-MNISTaverage accuracy (%): 91.09 -> 94.12Improved
raise rank increment per task from 1 to 4 (initial rank 11) (ablation)low-rank ITL · parameter count (P/R-MNIST)# parameters: 0.11M -> 0.2MHurt
update only the last 1 layer vs last 5 layers of ResNet18 (ablation)low-rank ITL · S-CIFAR100 (ResNet18)average accuracy (%): 34.38 -> 65.03Improved
use rank-1 ITL instead of Orthog Subspaceincremental task learning · S-miniImageNet (3-layer MLP)average accuracy (%): 38.1 -> 54.7Improved
use rank-1 ITL instead of HATincremental task learning · P-MNIST (3-layer MLP)average accuracy (%): 90.1 -> 85.6Hurt
use rank-1 ITL instead of Adam-NSCLincremental task learning · S-CIFAR-100 (ResNet18)average accuracy (%): 74.31 -> 68.46Hurt
use rank-1 ITL instead of Parallel 2 (similar parameter count)incremental task learning · S-CIFAR100 (3-layer MLP)average accuracy (%): 62.8 -> 65.9Improved

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