Research map / Architectures and efficiency / Pruning and efficient tuning
LQF: Linear Quadratic Fine-Tuning
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 LQF instead of NLFT | ResNet-50 fine-tuning · Stanford Dogs | test error: 13.8 -> 12.4 | Helped |
| Use LQF instead of NLFT | ResNet-50 fine-tuning · Caltech-256 | test error: 14.1 -> 14.5 | Hurt |
| Use LQF instead of NLFT | ResNet-50 fine-tuning · Chest X-Ray | test error: 4.6 -> 7.1 | Hurt |
| Use LQF instead of NLFT | ResNet-50 fine-tuning · Malaria Cells | test error: 3.2 -> 4.1 | Hurt |
| Use LQF instead of NLFT | ResNet-50 fine-tuning · Oxford Flowers | test error: 7.1 -> 7.1 | No clear effect |
| Replace global average pooling with square-root bilinear pooling (ablation) | LQF · CUB-200 | test error: 24.0 -> 17.1 | Helped |
| Replace global average pooling with square-root bilinear pooling (ablation) | LQF · Aircrafts | test error: 34.5 -> 29.2 | Helped |
| Replace global average pooling with square-root bilinear pooling (ablation) | LQF · Cars | test error: 27.1 -> 16.3 | Helped |
| Use mean-squared error instead of cross-entropy (ablation) | LQF · image classification datasets | test error | Helped |
| Add K-FAC pre-conditioning instead of using standard SGD (ablation) | LQF · image classification datasets | test error | Helped |
| Use Leaky-ReLU instead of ReLU (ablation) | LQF · image classification datasets | test error | Helped |
| Use MSE instead of cross-entropy (ablation) | NLFT · low-shot classification | test error | Helped |
Closest papers by meaning
- Low-Rank Rescaled Vision Transformer Fine-Tuning: A Residual Design Approach (CVPR 2024 · similarity 0.86)
- Tuning Stable Rank Shrinkage: Aiming at the Overlooked Structural Risk in Fine-tuning (CVPR 2024 · similarity 0.86)
- Directional Gradient Projection for Robust Fine-Tuning of Foundation Models (ICLR 2025 · similarity 0.86)
- QuadEnhancer: Leveraging Quadratic Transformations to Enhance Deep Neural Networks (NeurIPS 2025 · similarity 0.85)
- Linearization Explains Fine-Tuning in Large Language Models (NeurIPS 2025 · similarity 0.85)
- SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors (NeurIPS 2024 · similarity 0.85)
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.