发表机构
Bogazici University; University of Cambridge（博阿齐奇大学; 剑桥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究人机交互中社交适宜动作生成问题，提出模型无关框架StARS，结合协同过滤与可学习场景表示生成用户特定适宜性得分，在多数据集评估中提升性能与一致性，支持个性化动作选择。
AI 中文摘要
在人机交互中，社交适宜性并非普遍适用：不同人在相同情境下对同一机器人动作的评判可能不同。为捕捉这种主体间的变异性，我们将社交适宜动作生成重新表述为受推荐系统启发的偏好建模问题，把注释者视为用户，情境/场景视为物品，一组候选机器人动作的适宜性得分视为目标。我们提出了StARS，这是一个新颖的模型无关框架，它将协同过滤与可学习的场景表示相结合，以生成针对候选机器人动作的用户特定适宜性得分。StARS是模型无关的：它可与各种场景编码器和主干集成，无需重新设计基础模型即可实现个性化。我们在两个社交感知机器人数据集MannersDB+和SocNav1上评估了StARS，并分析了稀疏偏好反馈下的鲁棒性。在跨数据集和主干的情况下，StARS持续提高了性能以及与注释者的一致性，支持与用户规范一致的个性化动作选择。我们的代码可在该https网址公开获取。
英文摘要
Social appropriateness in human-robot interaction (HRI) is not universal: different people can judge the same robot action differently in the same situation. To capture this inter-subject variability, we reformulate socially appropriate action generation as a preference modelling problem inspired by recommender systems, treating annotators as users, contexts/scenes as items, and appropriateness scores over a set of candidate robot actions as targets. We propose StARS, a novel model-agnostic framework that integrates collaborative filtering with learnable scene representations to generate user-specific appropriateness scores over candidate robot actions. StARS is model-agnostic: it can be integrated with various scene encoders and backbones, enabling personalisation without redesigning the underlying model. We evaluate StARS on two socially aware robotics datasets, MannersDB+ and SocNav1, and analyse robustness under sparse preference feedback. Across datasets and backbones, StARS consistently improves performance and agreement with annotators, supporting personalised action selection aligned with user norms. Our code is publicly available at https://github.com/Cambridge-AFAR/StARS.git.
CommentsIROS 2026
