DinoCompanion: An Attachment-Theory Informed Multimodal Robot for Emotionally Responsive Child-AI Interaction

TL;DR

DinoCompanion is a novel attachment-theory-informed multimodal robot designed to support children's emotional development through developmentally appropriate, safe, and evaluable interactions, outperforming existing AI companions.

Contribution

It introduces a new attachment-theory grounded framework, a multimodal dataset, a risk-aware training method, and an evaluation benchmark for emotionally responsive child-AI interaction.

Findings

Achieved state-of-the-art attachment-related performance (57.15%).

Demonstrated secure base behaviors close to human experts (72.99%).

Validated the importance of multimodal fusion and uncertainty modeling.

Abstract

Children's emotional development fundamentally relies on secure attachment relationships, yet current AI companions lack the theoretical foundation to provide developmentally appropriate emotional support. We introduce DinoCompanion, the first attachment-theory-grounded multimodal robot for emotionally responsive child-AI interaction. We address three critical challenges in child-AI systems: the absence of developmentally-informed AI architectures, the need to balance engagement with safety, and the lack of standardized evaluation frameworks for attachment-based capabilities. Our contributions include: (i) a multimodal dataset of 128 caregiver-child dyads containing 125,382 annotated clips with paired preference-risk labels, (ii) CARPO (Child-Aware Risk-calibrated Preference Optimization), a novel training objective that maximizes engagement while applying epistemic-uncertainty-weighted risk penalties, and (iii) AttachSecure-Bench, a comprehensive evaluation benchmark covering ten attachment-centric competencies with strong expert consensus (\k{appa}=0.81). DinoCompanion achieves state-of-the-art performance (57.15%), outperforming GPT-4o (50.29%) and Claude-3.7-Sonnet (53.43%), with exceptional secure base behaviors (72.99%, approaching human expert levels of 78.4%) and superior attachment risk detection (69.73%). Ablations validate the critical importance of multimodal fusion, uncertainty-aware risk modeling, and hierarchical memory for coherent, emotionally attuned interactions.