Detecting Brittle Decisions for Free: Leveraging Margin Consistency in Deep Robust Classifiers
–Neural Information Processing Systems
First, we establish that margin consistency is a necessary and sufficient condition to use a model's logit margin as a score for identifying non-robust samples.
Neural Information Processing Systems
Oct-9-2025, 21:39:34 GMT
- Country:
- Asia
- India > Maharashtra
- Pune (0.04)
- Middle East > Jordan (0.04)
- India > Maharashtra
- Europe
- France > Provence-Alpes-Côte d'Azur (0.04)
- Germany > Baden-Württemberg
- Karlsruhe Region > Heidelberg (0.04)
- North America
- Canada
- Alberta > Census Division No. 15
- Improvement District No. 9 > Banff (0.04)
- Ontario > Toronto (0.14)
- Quebec (0.04)
- Alberta > Census Division No. 15
- United States > California
- San Diego County > San Diego (0.04)
- Canada
- Asia
- Genre:
- Research Report > Experimental Study (0.46)
- Industry:
- Information Technology (0.47)
- Technology: