Research map / Label-efficient and robust learning / Knowledge distillation
Learning To Retain While Acquiring: Combating Distribution-Shift in Adversarial Data-Free Knowledge Distillation
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 meta-learning-inspired acquisition-then-retention updates with a memory bank instead of MB-DFKD replay | Adversarial DFKD · SVHN | Accmax (%): 94.11 -> 95.88 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with generative replay instead of PRE-DFKD | Adversarial DFKD · SVHN | Accmax (%): 96.10 -> 96.15 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with a memory bank instead of MB-DFKD replay | Adversarial DFKD · CIFAR10 | Accmax (%): 93.03 -> 93.73 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with generative replay instead of PRE-DFKD | Adversarial DFKD · CIFAR10 | Accmax (%): 93.41 -> 94.02 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with a memory bank instead of MB-DFKD replay | Adversarial DFKD · CIFAR100 | Accmax (%): 76.14 -> 77.11 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with generative replay instead of PRE-DFKD | Adversarial DFKD · CIFAR100 | Accmax (%): 76.93 -> 77.21 | Helped |
| Use meta-learning-inspired acquisition-then-retention updates with a memory bank instead of MB-DFKD replay | Adversarial DFKD · Tiny-ImageNet | Accmax (%): 47.96 -> 47.96 | No clear effect |
| Use meta-learning-inspired acquisition-then-retention updates with generative replay instead of PRE-DFKD | Adversarial DFKD · Tiny-ImageNet | Accmax (%): 49.94 -> 49.88 | Hurt |
| Use meta-learning-inspired updates with memory-bank replay rather than naive memory-bank replay | Adversarial DFKD · SVHN | mean accuracy at >80th percentile: 81.67 -> 95.59 | Helped |
Closest papers by meaning
- Impartial Adversarial Distillation: Addressing Biased Data-Free Knowledge Distillation via Adaptive Constrained Optimization (AAAI 2024 · similarity 0.90)
- Hybrid Data-Free Knowledge Distillation (AAAI 2025 · similarity 0.90)
- Small Scale Data-Free Knowledge Distillation (CVPR 2024 · similarity 0.90)
- Teacher as a Lenient Expert: Teacher-Agnostic Data-Free Knowledge Distillation (AAAI 2024 · similarity 0.89)
- When Data-Free Knowledge Distillation Meets Non-Transferable Teacher: Escaping Out-of-Distribution Trap is All You Need (ICML 2025 · similarity 0.89)
- Robust Distillation via Untargeted and Targeted Intermediate Adversarial Samples (CVPR 2024 · similarity 0.87)
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.