AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Defending Unauthorized Model Merging via Dual-Stage Weight Protection

Wei-Jia Chen, Min-Yan Tsai, Cheng-Yi Lee, Chia-Mu Yu

CVPR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 1 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
apply the dual-stage protection before releasing the modelfine-tuned ViT-L-14 · 8 vision datasets (merged, TA)average accuracy: 84.47 -> 40.26 merged accuracyImproved
apply the protectionfine-tuned ViT-L-14 · 8 vision datasets (standalone)accuracy: 99.24 -> 98.25 (GTSRB)No clear effect
use the dual-stage method instead of PaRaMSmerged ViT-L-14 · 8 vision datasets (TA)average accuracy drop: 30.76 -> 52.11Improved
combine Stage 2 with Stage 1 (ablation)defense · GTSRB (merged)accuracy: 64.86 -> 12.91Improved
combine Stage 1 with Stage 2 (ablation)defense · MNIST (merged)accuracy: 13.37 -> 11.35Improved
combine Stage 1 with Stage 2 (ablation)defense · SUN397 (merged)accuracy: 76.38 -> 54.27Improved
Unmask adaptive attack (subtract protected model)defense · 8 vision datasets averagemerged accuracy: 40.26 -> 64.64Hurt
gradient-projection adaptive attackdefense · 8 vision datasets averagemerged accuracy: 40.26 -> 51.77Hurt
protection on diffusion model (SD1.5)defense vs PaRaMS · image generationvisual fidelity: no numbersMixed

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