Mind the Context: Continual Learning of Socially Appropriate Robot Actions via Environmental-Social Disentanglement

Mind the Context: Continual Learning of Socially Appropriate Robot Actions via Environmental-Social Disentanglement

|Authors: Rafal Robert Karpinski, Fethiye Irmak Dogan, Nikhil Churamani, Yiming Luo, Maartje M. A. de Graaf, Davide Dell’Anna, Hatice Gunes

|Venue: arXiv preprint | Year: 2026

|arXiv: 2608.13448v1

Abstract

Social robots are expected to operate across diverse environments, where similar arrangements can imply different socially appropriate actions, e.g., starting a conversation may be acceptable in a crowded home but disruptive in an office meeting. Because such norms and environments cannot all be anticipated in advance, robots require continual learning (CL) to adapt from sequential experience while retaining previously acquired knowledge. Prior work has studied CL for generating socially appropriate robot actions, but it has not addressed domain-incremental settings in which the robot incrementally encounters diverse contexts (e.g., living room, meeting room, office, hallway), where both environmental (e.g., whether the space is open or cluttered with furniture) and social cues (e.g., how people or other agents are positioned around the robot) jointly shape the appropriateness of robot actions. We address this gap with the Explicit Disentanglement Dual-Branch (EDD) framework. EDD explicitly separates environmental and social-agent related knowledge and uses replay-based rehearsal to mitigate forgetting while learning the appropriateness of robot actions (e.g., cleaning, serving, starting a conversation) across several indoor domains. Experiments show that EDD outperforms several state-of-the-art baselines, and ablation studies further evaluate different disentanglement strategies and the sensitivity to domain ordering. Our code is publicly available at https://github.com/Cambridge-AFAR/Mind-the-Context.git.

Key Contributions

• Social robots are expected to operate across diverse environments, where similar arrangements can…
• Because such norms and environments cannot all be anticipated in advance, robots require continua…
• Prior work has studied CL for generating socially appropriate robot actions, but it has not addre…
• We address this gap with the Explicit Disentanglement Dual-Branch (EDD) framework
• EDD explicitly separates environmental and social-agent related knowledge and uses replay-based r…

Topics

• human-robot-interaction

Code & Data

• GitHub repository: https://github.com/Cambridge-AFAR/Mind-the-Context.git

BibTeX

@article{Karpinski2026_260813448v1,
  title     = {Mind the Context: Continual Learning of Socially Appropriate Robot Actions via Environmental-Social Disentanglement},
  author    = {Rafal Robert Karpinski and Fethiye Irmak Dogan and Nikhil Churamani and Yiming Luo and Maartje M. A. de Graaf and Davide Dell'Anna and Hatice Gunes},
  year      = {2026},
  eprint    = {2608.13448v1},
  archivePrefix = {arXiv},
  primaryClass  = {cs.RO},
  url       = {https://arxiv.org/abs/2608.13448v1}
}