Adversarial Robustness Toolbox (ART)
EvalsTool (not an eval — tooling/reference entry)
Evasion, poisoning, extraction, inference attacks and defenses.
- Category
- Adversarial ML (classic)
- Insurance exposure
- Model/weight theft
- Publisher
- IBM / LF AI
- Publisher type
- Big tech
- License
- MIT
- Status
- Active (<90d)
- Last commit
- 2026-10-08
- Stars
- 6279
- Notes
- Includes model-extraction attacks.
Wanich Risk Lab, 2026.