
Rubrics-as-an-Attack-Surface
Research code for Rubric-Induced Preference Drift (RIPD): evolutionary rubric search, benchmark-preserving selection, and DPO policy misalignment…
adversarial-attackai-securityeducation+2
7

Research code for Rubric-Induced Preference Drift (RIPD): evolutionary rubric search, benchmark-preserving selection, and DPO policy misalignment…

Research implementation of an LLM jailbreak agent that bypasses commercial model guardrails using harmless prompt weaving and adaptive tree search.

Alignment-research scaffold (autoresearch-style) for LLM guardrails: search over a single policy.md surface