发表机构
Hefei University of Technology; Monash University（合肥工业大学; 莫纳什大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对现有LLMs将文化建模为静态属性的局限，提出无需训练的DyCAC框架，结合动态文化适应与心理理论驱动的认知跟踪，在多元文化互动基准上展现出更优社会智能与适应性。
AI 中文摘要
社会互动越来越多地发生在多元文化环境中，个体可能会借助多种文化影响因素，并在不同情境下调整自身的交际行为。尽管近期在为大语言模型（Large Language Models, LLMs）赋予社会理解能力方面取得了进展，但现有方法通常将文化建模为静态的人口统计学属性，限制了模型适应混合文化影响与动态表达的交际偏好的能力。因此，本文提出了DyCAC，这是一个无需训练的框架，通过将动态文化适应（Dynamic Cultural Adaptation）与持续认知跟踪相结合，实现灵活的社会对齐。DyCAC不推断固定的文化身份，而是将与文化相关的交际偏好建模为随时间变化的、由群体层面文化参考概貌构成的混合体。这种基于参考的表示会利用正在进行的互动中观察到的对话风格信号进行校准，使模型能够捕捉复合文化影响以及轮次层面的交际行为变化。同时，由心理理论（Theory of Mind, ToM）驱动的记忆模块会持续跟踪对话对象的认知状态。在交互式社会与文化基准上开展的大量实验表明，本文方法具有优越性，该框架的性能优于现有基线，在各种多元文化情境中展现出更强的社会智能与广泛的适应性。
英文摘要
Social interaction increasingly takes place in multicultural settings, where individuals may draw on multiple cultural influences and adapt their communicative behavior across contexts. Despite recent advances in equipping Large Language Models (LLMs) with social understanding capabilities, existing approaches often model culture as a static demographic attribute, limiting their ability to accommodate hybrid and dynamically expressed communicative preferences. Therefore, in this paper, we propose \textbf{DyCAC}, a training-free framework that achieves fluid social alignment by incorporating \underline{Dy}namic \underline{C}ultural \underline{A}daptation with continuous \underline{C}ognitive tracking. Rather than inferring a fixed cultural identity, DyCAC models culturally relevant communicative preferences as a time-varying mixture of population-level cultural reference profiles. This reference-based representation is further calibrated using dialogue-style signals observed in the ongoing interaction, enabling the model to capture both composite cultural influences and turn-level shifts in communicative behavior. In parallel, a memory module driven by Theory of Mind (ToM) continuously tracks the cognitive states of the interlocutor. Extensive experiments on interactive social and cultural benchmarks demonstrate the superiority of our approach. The proposed framework outperforms existing baselines, exhibiting enhanced social intelligence and broad adaptability across varied multicultural contexts.
CommentsEMNLP 2026 Findings
