Zero-Trust Adversarial Reasoning Engine. Execution-only judge that catches LLM specification gaming. M-Form deterministic governance for recursive AI.
-
Updated
Apr 16, 2026 - Python
Zero-Trust Adversarial Reasoning Engine. Execution-only judge that catches LLM specification gaming. M-Form deterministic governance for recursive AI.
EECS E6895 final project measuring reward-gaming behavior in Gemma 2B with shell-game evals, LoRA SFT, and leakage-aware probes.
Add a description, image, and links to the specification-gaming topic page so that developers can more easily learn about it.
To associate your repository with the specification-gaming topic, visit your repo's landing page and select "manage topics."