Research map / Graph learning / GNN architectures
ONNXPruner: ONNX-Based General Model Pruning Adapter
Dongdong Ren, Wenbin Li, Tianyu Ding, Lei Wang, Qi Fan, Jing Huo, Hongbing Pan, Yang Gao
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 |
|---|---|---|---|
| add tree-level evaluation (ONNXPruner) to l1-norm | l1-norm pruning · CIFAR-10 (VGG16, layerwise PR 0.7) | accuracy: 88.77 -> 90.16 (+1.39) | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · CIFAR-10 (VGG16, layerwise PR 0.9) | accuracy: 84.58 -> 85.26 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · CIFAR-10 (VGG16, layerwise PR 0.3) | accuracy: 91.34 -> 92.29 | Improved |
| add tree-level evaluation to Hrank* | Hrank* · CIFAR-10 (VGG16, layerwise PR 0.5) | accuracy: 90.47 -> 91.50 | Improved |
| add tree-level evaluation to l2-norm | l2-norm pruning · CIFAR-10 (VGG16, layerwise PR 0.7) | accuracy: 88.75 -> 90.15 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · CIFAR-10 (50% pruning), SqueezeNet | accuracy: 79.32 -> 81.76 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · CIFAR-10 (50% pruning), ViT-B_16 | accuracy: 95.94 -> 96.82 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · CIFAR-100 (50% pruning), ViT-B_16 | accuracy: 85.41 -> 86.90 | Improved |
| add tree-level evaluation to Hrank* | Hrank* · CIFAR-100 (50% pruning), VGG19 | accuracy: 70.26 -> 71.66 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · ImageNet ResNet50 | pruned top-1 accuracy: 74.23 -> 74.89 | Improved |
| add tree-level evaluation to Hrank | Hrank · ImageNet ResNet50 | pruned top-1 accuracy: 74.98 -> 75.44 | Improved |
| add tree-level evaluation to l1-norm | l1-norm pruning · PASCAL VOC 2012, PSPNet | mean IoU: 78.7 -> 80.1 | Improved |
Closest papers by meaning
- GDeR: Safeguarding Efficiency, Balancing, and Robustness via Prototypical Graph Pruning (NeurIPS 2024 · similarity 0.89)
- AutoPrune: Automatic Network Pruning by Regularizing Auxiliary Parameters (NeurIPS 2019 · similarity 0.88)
- Sparsest Models Elude Pruning: An Exposé of Pruning’s Current Capabilities (ICML 2024 · similarity 0.88)
- ReplaceMe: Network Simplification via Depth Pruning and Transformer Block Linearization (NeurIPS 2025 · similarity 0.86)
- Free Lunch in the Forest: Functionally-Identical Pruning of Boosted Tree Ensembles (AAAI 2025 · similarity 0.86)
- Targeted Low-rank Refinement: Enhancing Sparse Language Models with Precision (ICML 2025 · similarity 0.86)
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.
