Learning social norms enhances compatibility in dynamic human-AI coordination

📄 arXiv: 2607.07021v1 📥 PDF

作者: Yi Yang, Siyuan Liu, Xin Gao, Huamu Sun, Chao Liu, Qing Zhou, Bingbing Nie

分类: cs.AI, cs.HC

发布日期: 2026-07-08

备注: 44 pages, 5 figures, supplementary information included


💡 一句话要点

通过学习社会规范提升动态人机协调的兼容性

🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)

关键词: 人机交互 社会规范 动态协调 人工智能 大型语言模型 行为预测 价值对齐

📋 核心要点

  1. 现有方法未能有效量化人类隐含的社会规范,导致AI在动态交互中与人类协调不佳。
  2. 本文提出通过识别和量化人类社会规范的原则,来提升AI代理与人类的协调能力。
  3. 实验结果显示,社会规范指导的AI在与人类的互动中得分显著提升,表现优于传统方法和人际交互。

📝 摘要(中文)

人类在动态交互中通过隐含的社会规范进行协调,这些规范作为互动代理之间的共享期望。随着AI代理(如大型语言模型)日益融入日常生活,它们在这些交互中扮演着重要角色。然而,现有方法未能有效量化这些规范,导致AI与人类的协调不够自然。本文通过对行人-车辆交互的实验研究,识别出三条人类社会规范的原则,并将其融入AI代理中,显著提升了人机协调的效果。实验结果表明,社会规范指导的AI在与人类的闭环交互任务中,得分几乎是基线策略的四倍,且超越人际交互43%。

🔬 方法详解

问题定义:本文旨在解决AI代理在动态人机交互中协调不佳的问题。现有方法主要依赖人类示范,但未能明确量化生成这些行为的社会规范,导致AI的表现不够自然和有效。

核心思路:论文的核心思路是识别和量化人类社会规范的原则,并将其融入AI代理的决策过程中,以提升其与人类的协调能力。通过明确的规范指导,AI能够更好地理解人类的期望和行为。

技术框架:研究构建了一个简化的实验平台,模拟行人-车辆交互的动态特征。通过收集3,456个动态人类交互数据,识别出三条社会规范原则:结果可预测性、价值对齐和优势意识。AI代理在决策时结合这些原则,从而改善协调效果。

关键创新:最重要的技术创新在于将隐含的社会规范形式化为明确的、可量化的原则。这一方法与现有依赖人类示范的策略本质不同,强调了规范在动态交互中的重要性。

关键设计:在模型设计中,采用了特定的损失函数来优化AI代理的决策过程,使其更好地遵循识别出的社会规范。同时,调整了网络结构以增强模型对动态交互的适应性。

🖼️ 关键图片

img_0
img_1

📊 实验亮点

实验结果显示,社会规范指导的AI在闭环交互任务中得分几乎是基线策略的四倍,且超越人际交互43%。这一显著提升表明,量化社会规范对改善人机协调具有重要意义。

🎯 应用场景

该研究的潜在应用领域包括自动驾驶、智能助手和人机协作等场景。通过提升AI在动态交互中的协调能力,可以实现更自然的人机互动,增强用户体验,并推动AI技术在社会中的更广泛应用。

📄 摘要(原文)

Humans continuously coordinate with others in dynamic interactions, often through implicit, hard-to-quantify social norms that act as shared tacit expectations among interacting agents. As AI agents, including large language models (LLMs), become embedded in daily life, they increasingly participate in such interactions and reshape social interaction structures. Yet they often fail to coordinate with humans in an effective, considerate, and natural manner. We hypothesize that this gap arises because existing approaches align model behavior with human demonstrations without explicitly quantifying the underlying norms that generate such behavior. We selected pedestrian-vehicle interaction as a representative dynamic interaction and developed a simplified experimental platform that captures its key interactive features. From 3,456 dynamic human interactions collected via this platform, we identified three principles underlying human social norms: outcome predictability, value alignment, and advantage awareness. Incorporating these principles into AI agents significantly improves human-AI coordination. In the closed-loop interaction task with humans, the social-norm-informed LLM achieved a nearly fourfold higher total score than the baseline strategy and outperformed human-human interactions by 43%. These findings indicate that formalizing tacit social norms into explicit, quantifiable principles can enable AI agents to achieve mutually beneficial coordination in dynamic interactions, supporting their more natural integration into human society.