Understanding Moral Reasoning Trajectories in Large Language Models: Toward Probing-Based Explainability

Excerpt

arXiv:2603.16017v2 Announce Type: replace Abstract: Large language models (LLMs) increasingly participate in morally sensitive decision-making, yet how they organize ethical frameworks across reasoning steps remains underexplored. We introduce moral reasoning trajectories, sequences of ethical framework invocations across intermediate reasoning steps, and analyze their dynamics across six models and three benchmarks. We find that moral reasoning involves systematic multi-framework deliberation: