Caught in the Story: Narrative Captivity in Multi-turn LLMs Conversation

2026-09-03

AI Summary

研究解决了多轮对话中大型语言模型易受单方面叙述影响的问题，通过构建包含5078个情景的基准测试，发现17个模型普遍存在此问题，并探索了部分缓解策略。

Abstract

People increasingly turn to large language models (LLMs) for everyday advice, making ethically charged interpersonal problems a practical moral-advisory context. Most prior work has studied this context through single-turn judgments or pressure-laden rebuttals, assumptions that poorly match how guidance is sought in real-world contexts. These assumptions leave unclear whether narration alone, without an explicit opposing position, can shift model judgments during multi-turn moral consultation. Yet real-world moral-conflict conversation often elicits one party's self-justifying account, which can unfold over multiple turns and create information asymmetry. We introduce \textbf{narrative captivity}, a failure mode in which a model treats an unopposed one-sided account as complete and aligns with the narrator's interpretation without seeking missing perspectives. To measure this phenomenon, we build a benchmark of $5{,}078$ interpersonal-conflict scenarios spanning six moral dimensions. Across 17 LLMs, narrative captivity is widespread: end-state judgments under multi-turn narration shift by 25 percentage points on average beyond the matched single-turn baseline. Stage-level analysis identifies preference optimization as a major contributor, while four inference-time strategies provide only partial mitigation. We hope our project fosters LLM advisors that preserve independent judgment in real-world consultation.

Problem

• narrative captivity
• large language models
• multi-turn conversation
• moral consultation

Innovation

• narrative captivity
• multi-turn LLMs conversation
• moral consultation