Research map / Label-efficient and robust learning / Domain adaptation and generalization
Partial Network Cloning
Read the paperCheck 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 full PNC instead of Continual+KD | LeNet5 · MNIST Avg.-S | accuracy: 91.7 -> 97.0 | Helped |
| Use full PNC instead of Continual+KD | ResNet-18 · CIFAR-10 Avg.-S | accuracy: 69.9 -> 95.0 | Helped |
| Use full PNC instead of Continual+KD | ResNet-50 · CIFAR-100 Avg.-S | accuracy: 66.7 -> 76.4 | Helped |
| Use full PNC instead of Continual+KD | ResNet-18 · Tiny-ImageNet Avg.-S | accuracy: 57.1 -> 62.8 | Helped |
| Add locality-based module localization (ablation) | PNC · MNIST Avg.-S | accuracy: 94.0 -> 97.0 | Helped |
| Add adaptive insertion (ablation) | PNC · MNIST Avg.-S | accuracy: 93.9 -> 97.0 | Helped |
| Add locality-based module localization (ablation) | PNC · CIFAR-10 Avg.-S | accuracy: 91.7 -> 95.0 | Helped |
| Add adaptive insertion (ablation) | PNC · CIFAR-10 Avg.-S | accuracy: 90.3 -> 95.0 | Helped |
| Add locality-based module localization (ablation) | PNC · CIFAR-100 Avg.-S | accuracy: 71.9 -> 76.4 | Helped |
| Add adaptive insertion (ablation) | PNC · CIFAR-100 Avg.-S | accuracy: 65.4 -> 76.4 | Helped |
| Add locality-based module localization (ablation) | PNC · Tiny-ImageNet Avg.-S | accuracy: 63.4 -> 62.8 | Hurt |
| Add adaptive insertion (ablation) | PNC · Tiny-ImageNet Avg.-S | accuracy: 61.2 -> 62.8 | Helped |
Closest papers by meaning
- A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone (NeurIPS 2025 · similarity 0.83)
- Progressive Network Grafting for Few-Shot Knowledge Distillation (AAAI 2021 · similarity 0.82)
- MergeNet: Knowledge Migration Across Heterogeneous Models, Tasks, and Modalities (AAAI 2025 · similarity 0.81)
- Fantastic Gains and Where to Find Them: On the Existence and Prospect of General Knowledge Transfer between Any Pretrained Model (ICLR 2024 · similarity 0.81)
- Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations (ICML 2025 · similarity 0.81)
- KINDLE: Knowledge-Guided Distillation for Prior-Free Gene Regulatory Network Inference (NeurIPS 2025 · similarity 0.80)
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.