failurefirst/failure-first

Adversarial evaluation framework for embodied and agentic AI — failure-first methodology, jailbreak corpus, VLA red-teaming, and policy research.

语言：Astro · GitHub 星标：4 · Fork：0

话题：adversarial-ml · agentic-ai · ai-safety · alignment · benchmark · embodied-ai · failure-analysis · jailbreak

在 GitHub 上查看原仓库