Reproduce “Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.