Research map / Adversarial robustness and security / Certified and empirical robustness
Efficient Adversarial Contrastive Learning via Robustness-Aware Coreset Selection
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 RCS instead of random selection | ACL pre-trained on ImageNet-1K with WRN-28-10 · ImageNet-1K to CIFAR-10 | robust accuracy (ALF, RA): 27.14 -> 31.13 | Helped |
| Use RCS instead of random selection | ACL pre-trained on ImageNet-1K with WRN-28-10 · ImageNet-1K to CIFAR-100 | robust accuracy (ALF, RA): 11.93 -> 14.89 | Helped |
| Use RCS instead of random selection | ACL pre-trained on ImageNet-1K with WRN-28-10 · ImageNet-1K to CIFAR-10 | robust accuracy (AFF, RA): 50.12 -> 52.21 | Helped |
| Use RCS instead of random selection | ACL pre-trained on ImageNet-1K with WRN-28-10 · ImageNet-1K to CIFAR-100 | robust accuracy (AFF, RA): 25.46 -> 28.24 | Helped |
| Use RCS instead of random selection | ACL with ResNet-18 · STL-10 | robust accuracy (SLF, RA), subset fraction 0.05: 22.00 -> 26.99 | Helped |
| Use RCS instead of random selection | ACL with ResNet-18 · STL-10 | robust accuracy (SLF, RA), subset fraction 0.1: 24.16 -> 28.66 | Helped |
| Use RCS instead of random selection | ACL with ResNet-18 · STL-10 | robust accuracy (SLF, RA), subset fraction 0.2: 27.09 -> 30.09 | Helped |
| Use RCS instead of random selection | DynACL with ResNet-18 · STL-10 | robust accuracy (SLF, RA), subset fraction 0.1: 30.72 -> 40.88 | Helped |
| Use RCS rather than training on the entire set | ACL with WRN-28-10 · ImageNet-1K pre-training | running time (hours): 650.2 -> 111.8 | Helped |
| Use RCS rather than standard CL | ImageNet-1K pre-training · ImageNet-1K pre-training | running time (hours): 147.4 -> 111.8 | Hurt |
Closest papers by meaning
- Robust Pre-Training by Adversarial Contrastive Learning (NeurIPS 2020 · similarity 0.90)
- When does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning? (NeurIPS 2021 · similarity 0.89)
- Weakly Supervised Contrastive Adversarial Training for Learning Robust Features from Semi-supervised Data (CVPR 2025 · similarity 0.86)
- Boosting Adversarial Robustness with CLAT: Criticality Leveraged Adversarial Training (ICML 2025 · similarity 0.86)
- Robust NAS under adversarial training: benchmark, theory, and beyond (ICLR 2024 · similarity 0.85)
- Confidence-aware Contrastive Learning for Selective Classification (ICML 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.