AI Paper Atlas

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

Practical Continual Forgetting for Pre-Trained Vision Models

Hongbo Zhao, Fei Zhu, Bolin Ni, Feng Zhu, Gaofeng Meng, Zhaoxiang Zhang

TPAMI 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Synthetic data and merging for CLIP · 6 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 prototype regularization (GS-LoRA to GS-LoRA++)GS-LoRA · CASIA-Face100 4-shot single-step 100-10H-Mean: 63.61 -> 69.47Improved
add prototype regularization (GS-LoRA to GS-LoRA++)GS-LoRA · CASIA-Face100 4-shot single-step 100-90H-Mean: 26.82 -> 60.86Improved
add prototype regularization (GS-LoRA to GS-LoRA++)GS-LoRA · CASIA-Face100 4-shot continual 100-20H-Mean: 66.02 -> 71.24Improved
add prototype regularization with full dataGS-LoRA · CASIA-Face100 single-step 100-10H-Mean: 71.76 -> 72.04No clear effect
add group sparse loss (alpha 0 -> 0.01) (ablation)GS-LoRA++ · CASIA-Face100 4-shot, forget 10H-Mean: 66.74 -> 69.47Improved
raise sparse intensity too far (alpha 0.1 -> 0.5) (ablation)GS-LoRA++ · CASIA-Face100 4-shot, forget 10H-Mean: 71.20 -> 66.98Hurt
increase LoRA rank from 2 to 8 (ablation)GS-LoRA++ · Deformable DETR forget 10 classesH-Mean: 37.98 -> 42.38Improved
increase LoRA rank from 8 to 16 (ablation)GS-LoRA++ · Deformable DETR forget 10 classesH-Mean: 42.38 -> 42.53No clear effect
increase shots from 1 to 4 (ablation)GS-LoRA++ · CASIA-Face100 few-shot, forget 10H-Mean: 32.14 -> 69.74Improved
add random augmentation when computing prototypes (1-shot) (ablation)GS-LoRA++ · CASIA-Face100 1-shotH-Mean: 32.14 -> 42.64Improved
add random augmentation when computing prototypes (2-shot) (ablation)GS-LoRA++ · CASIA-Face100 2-shotH-Mean: 53.72 -> 57.74Improved
cut forgetting data ratio from 0.05 to 0.02 (ablation)GS-LoRA++ · CASIA-Face100 data efficiencyH-Mean: 72.22 -> 67.82Hurt

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