CybORG CAGE-2

CybORG CAGE-2 is a cyber defense environment modeled as a Partially Observable Markov Decision Process (POMDP) used to evaluate the performance and cost-effectiveness of compound LLM agent designs in adversarial scenarios.

Papers using CybORG CAGE-2 (2)

Context, Reasoning, and Hierarchy: A Cost-Performance Study of Compound LLM Agent Design in an Adversarial POMDP2026FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast2026