Narration-of-Thought: Inference-Time Scaffolding for Defeasible Ethical Reasoning in Large Language Models
We introduce narration-of-thought (NoT), a system prompt that structures chain-of-thought into five sections: protagonist, stakeholders, two-step consequences, uncertainty, then commitment. NoT adds no training, parameters, or fine-tuning. On 100 DailyDilemmas scenarios across four generators from three vendors, NoT cuts stakeholder collapse from up to 31% to under 1% and uncertainty suppression from up to 72% to 1-24% on every model.
- ▪We introduce narration-of-thought (NoT), a system prompt that structures chain-of-thought into five sections: protagonist, stakeholders, two-step consequences, uncertainty, then commitment.
- ▪NoT adds no training, parameters, or fine-tuning.
- ▪On 100 DailyDilemmas scenarios across four generators from three vendors, NoT cuts stakeholder collapse from up to 31% to under 1% and uncertainty suppression from up to 72% to 1-24% on every model.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2606.26366 (cs) [Submitted on 24 Jun 2026] Title:Narration-of-Thought: Inference-Time Scaffolding for Defeasible Ethical Reasoning in Large Language Models Authors:Patrick Cooper, Alvaro Velasquez View a PDF of the paper titled Narration-of-Thought: Inference-Time Scaffolding for Defeasible Ethical Reasoning in Large Language Models, by Patrick Cooper and 1 other authors View PDF HTML (experimental) Abstract:Standard chain-of-thought on moral dilemmas exhibits two failure modes: stakeholder collapse (the trace names at most one party with a stake in the outcome) and uncertainty suppression (no explicit unknowns or hedges before committing to an action).
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
