AI Paper Atlas

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

SMoLoRA: Exploring and Defying Dual Catastrophic Forgetting in Continual Visual Instruction Tuning

Ziqi Wang, Chang Che, Qi Wang, Yangyang Li, Zenglin Shi, Meng Wang

ICCV 2025 · Continual learning › Continual fine-tuning of LLMs · approach: Mixture-of-experts for multimodal continual tuning · 23 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 visual understanding (VU) module (ablation)SeqLoRA · CVIT upstream single-typeAP: 46.21 -> 53.49Improved
add instruction following (IF) module (ablation)SeqLoRA · CVIT upstream single-typeAP: 46.21 -> 71.97Improved
add instruction following (IF) module (ablation)SeqLoRA · CVIT upstream single-typeMIF: 78.35 -> 98.38Improved
combine VU and IF by averaging outputs (ablation)SeqLoRA · CVIT upstream single-typeAP: 71.97 -> 75.16Improved
add adaptive fusion (AF) module (ablation)SMoLoRA (VU+IF) · CVIT upstream single-typeAP: 75.16 -> 83.44Improved
add adaptive fusion (AF) module (ablation)SMoLoRA (VU+IF) · CVIT upstream single-typeBWT: -10.99 -> -3.23Improved
add adaptive fusion (AF) module (ablation)SMoLoRA (VU+IF) · CVIT upstream single-typeMIF: 97.43 -> 97.79Improved
increase number of LoRA blocks beyond 8 or 16 (ablation)SMoLoRA · CVIT upstream single-typeall metricsNo clear effect
increase number of LoRA blocks from small to 4-8 (ablation)SMoLoRA · CVIT upstream single-typeall metricsImproved
SMoLoRA versus MoeLoRA on upstream tasksMoeLoRA · CVIT upstream single-typeAP: 45.27 -> 83.44Improved
SMoLoRA versus second-best on zero-shot unseen tasks (AP)DoRA · CVIT zero-shot downstream (single-type)AP: 25.21 -> 34.35Improved
SMoLoRA versus second-best few-shotDoRA · CVIT Places365 few-shot (single-type)AP: 29.42 -> 40.32Improved

Later papers that compared with this method

1 later accepted paper reports SMoLoRA as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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