AI Paper Atlas

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

Shared LoRA Subspaces for almost Strict Continual Learning

Prakhar Kaushik, Ankit Vaidya, Shravan Sunil Chaudhari, Rama Chellappa, Alan Yuille

ECCV 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation · 3 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
finetune task coefficients after merging (Share-full)Share (CL) · Continual GLUE averageaverage score: 78.69 -> 83.44Improved
continual Share instead of per-task LoRALoRA (non-CL) · Continual GLUE averageaverage score: 83.43 -> 78.69Hurt
increase temporary factors phi from 1 to 3 (ablation)Share-full · Continual GLUE RTEaccuracy: 69.675 -> 74.368Improved
increase temporary factors phi from 3 to 4 (ablation)Share-full · Continual GLUE RTEaccuracy: 74.368 -> 73.285No clear effect
increase pseudo-rank p (ablation)Share-full · Continual GLUE RTEaccuracyImproved
increase number of principal factors k (ablation)Share-full · MRPCaccuracyNo clear effect
share low-rank subspace across tasks, backward transferShare · Continual GLUE CoLAMatthews correlation: 56.00 -> 59.81Improved
Share instead of TIES mergingTIES (non-CL) · OOD LoRA tasks (Mistral-7B)Rouge-L: 21.12 -> 55.89Improved
Share instead of DAPDAP · CIFAR-100 (10 tasks)final accuracy: 94.05 -> 94.20No clear effect
Share (replay-free) instead of iNeMO (replay)iNeMO · Pascal3D+pi/6 accuracy: 79.28 -> 81.80Improved
Share (replay-free) instead of iNeMO (replay)iNeMO · Occluded Pascal3D+ L3pi/6 accuracy: 34.01 -> 35.50Improved
Share instead of non-CL LoRA for sequential text-to-image tasksLoRA (non-CL) · Text-to-image T1 (Flux)CLIP score: 22.776 -> 23.382Improved

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