Research map / Architectures and efficiency / Pruning and efficient tuning
Parameter-Efficient Model Adaptation for Vision Transformers
Xuehai He, Chunyuan Li, Pengchuan Zhang, Jianwei Yang, Xin Eric Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use KAdaptation rather than the strongest efficient baseline, LoRA | CLIP ViT-B-224/32 · ELEVATER 20 datasets, 5-shot | average accuracy (%): 66.32 -> 68.92 | Improved |
| use KAdaptation rather than LoRA | CLIP · seven full-shot image-classification datasets | average accuracy (%): 79.3 -> 79.6 | Improved |
| use KAdaptation rather than LoRA | Supervised ViT · seven full-shot image-classification datasets | average accuracy (%): 78.6 -> 79.2 | Improved |
| use KAdaptation instead of full-model fine-tuning | CLIP · seven full-shot image-classification datasets | average accuracy (%): 80.6 -> 79.6 | Hurt |
| apply KAdaptation to MLP modules instead of attention modules (ablation) | KAdaptation · CIFAR10, CIFAR100, SUN397 | average accuracy (%): 88.1 -> 86.6 | Hurt |
| place Adapters in attention layers rather than using standard Adapter placement (ablation) | Adapter-tuning · CIFAR10, CIFAR100, SUN397 | average accuracy (%): 87.7 -> 54.1 | Hurt |
| select attention modules based on their lower local intrinsic dimension | Supervised ViT · CIFAR100, first-layer component intrinsic-dimension experiment | local intrinsic dimension: 575 -> 300 | Improved |
Closest papers by meaning
- Time- Memory- and Parameter-Efficient Visual Adaptation (CVPR 2024 · similarity 0.90)
- Dynamic Tuning Towards Parameter and Inference Efficiency for ViT Adaptation (NeurIPS 2024 · similarity 0.89)
- Learning Where to Edit Vision Transformers (NeurIPS 2024 · similarity 0.88)
- Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation (NeurIPS 2024 · similarity 0.88)
- Split Adaptation for Pre-trained Vision Transformers (CVPR 2025 · similarity 0.88)
- PELA: Learning Parameter-Efficient Models with Low-Rank Approximation (CVPR 2024 · similarity 0.88)
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.
