Simple Role Assignment is Extraordinarily Effective for Safety Alignment

ACL ARR 2026 January Submission10037 Authors

Keywords: value, alignment, role, llm, judge, reward, safety, prompt, feedback

Abstract: Principle-based alignment often lacks context sensitivity and completeness. Grounded in Theory of Mind, we propose role conditioning as a compact alternative: social roles (e.g., mother, judge) implicitly encode both values and the cognitive schemas required to apply them. We introduce a training-free pipeline featuring a role-conditioned generator and iterative role-based critics for refinement. Across five model families, our approach consistently outperforms principle-based, Chain-of-Thought (CoT) and other baselines across benchmarks. Notably, it reduces unsafe outputs on the WildJailbreak benchmark from 81.4\% to 3.6\% with DeepSeek-V3. Not only for common safety benchmarks, it consistently applies for agentic safety tasks. These results establish role assignment as a powerful, interpretable paradigm for AI alignment and LLM-as-a-Judge construction.

Paper Type: Long

Research Area: Safety and Alignment in LLMs

Research Area Keywords: safety and alignment for agents,

Contribution Types: NLP engineering experiment, Theory

Languages Studied: English

Submission Number: 10037