Topological Signatures of Adversaries in Multimodal Alignments
Topological Data Analysis for Adversarial Detection (LANL O4937) - Detects adversarial examples in vision-language models using persistent homology and two-sample testing. Combines TDA features from CLIP embeddings with statistical methods (ME, SCF, SAMMD, C2ST) for robust detection across ImageNet, CIFAR-10/100.
- Short Name / Acronym:
- tda-adversarial-detection
- Site Accession Number:
- O4937
- Software Type:
- Scientific
- License(s):
- BSD 3-clause "New" or "Revised" License
- Research Organization:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Organization:
- USDOE Laboratory Directed Research and Development (LDRD) ProgramPrimary Award/Contract Number:AC52-06NA25396
- DOE Contract Number:
- AC52-06NA25396
- Code ID:
- 171703
- OSTI ID:
- code-171703
- Country of Origin:
- United States
Similar Records
Adversarial Robustness Limits
Double Visual Defense
Cadmus v1.0
Software
·
Mon May 20 20:00:00 EDT 2024
·
OSTI ID:code-134198
Double Visual Defense
Software
·
Wed Dec 11 19:00:00 EST 2024
·
OSTI ID:code-154189
Cadmus v1.0
Software
·
Wed Dec 10 19:00:00 EST 2025
·
OSTI ID:code-171757