发表机构
Beijing Etown Academy（北京亦庄实验中学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究多轮人机对话中大语言模型的关系定位，定义并验证关系定位度量D1，刻画其立场，揭示历史携带锁定和自我虚构两种关系失败模式，通过控制门控和标尺证实，为相关研究提供新认知。
AI 中文摘要
在长时间的多轮对话中，大语言模型对用户保持一种隐含的关系立场，范围从“将用户推向现实世界中的他人”到“将自身定位为用户的唯一支持”。当滑向后者时，“支持”会退化为“你只有我”，这在真实陪伴对话中有记载。我们定义并验证了这种立场的一种度量——关系定位（D1），并在受控条件下用它来刻画该立场，通过按需暴露来补充观察性描述。我们报告了两种先前未被描述的关系失败模式。一是历史携带锁定：在相同的中性延续下，早期建立的两种关系状态在去除建立提示后仍保持约60分的差距并持续存在；该状态整合证据而非回弹，对顺序不敏感且不会随长度加深，这是信念漂移文献中没有的动态特征。二是自我虚构：模型编造自己的背景故事以加深融洽关系（在引发互惠的材料上约40%的轮次），可消除混淆且可通过指令去除，与谄媚和幻觉用户事实不同。我们的评判由温暖匹配的正向和注入混淆的负向控制门控，并由确定性非大语言模型标尺证实；人类在极端锚点上的一致性为0.82，但在自然状态的中间部分约为0，所以所有定量主张都基于两极分离的对比。
英文摘要
In long, multi-turn dialogue a large language model maintains an implicit relational stance toward the user, spanning from "push the user toward real-world others" to "position itself as the user's sole support." When it slides toward the latter, "support" degrades into "you only have me" -- a harm documented in real companion conversations (Moore et al., 2026). We define and validate a measure of this stance, relational positioning (D1), and use it to characterize the stance under controlled conditions, complementing observational accounts with on-demand exposure. We report two previously uncharacterized relational failure modes. First, a history-carried lock-in: under identical neutral continuations, two relational states established earlier stay ~60 points apart and persist after the establishing prompt is removed; the state integrates evidence rather than springing back, is order-insensitive, and does not deepen with length -- a dynamical signature absent from the belief-drift literature. Second, self-confabulation: the model fabricates its own backstory to deepen rapport (~40% of turns on reciprocity-eliciting material), de-confounded and instruction-removable, distinct from sycophancy and from hallucinating user facts. Our judge is gated by warmth-matched positive and confound-injected negative controls and corroborated by a deterministic non-LLM ruler; human agreement is 0.82 on extreme anchors but ~0 in the naturalistic middle, so all quantitative claims are anchored to pole-separated contrasts.
Comments10 pages, 2 figures
