Designing Social Robots with LLMs for Engaging Human Interaction

Large Language Models (LLMs), particularly those enhanced through Reinforcement Learning from Human Feedback, such as ChatGPT, have opened up new possibilities for natural and open-ended spoken interaction in social robotics. However, these models are not inherently designed for embodied, multimodal...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria Pinto-Bernal, Matthijs Biondina, Tony Belpaeme
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/11/6377
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850160506293190656
author Maria Pinto-Bernal
Matthijs Biondina
Tony Belpaeme
author_facet Maria Pinto-Bernal
Matthijs Biondina
Tony Belpaeme
author_sort Maria Pinto-Bernal
collection DOAJ
description Large Language Models (LLMs), particularly those enhanced through Reinforcement Learning from Human Feedback, such as ChatGPT, have opened up new possibilities for natural and open-ended spoken interaction in social robotics. However, these models are not inherently designed for embodied, multimodal contexts. This paper presents a user-centred approach to integrating an LLM into a humanoid robot, designed to engage in fluid, context-aware conversation with socially isolated older adults. We describe our system architecture, which combines real-time speech processing, layered memory summarisation, persona conditioning, and multilingual voice adaptation to support personalised, socially appropriate interactions. Through iterative development and evaluation, including in-home exploratory trials with older adults (<i>n</i> = 7) and a preliminary study with young adults (<i>n</i> = 43), we investigated the technical and experiential challenges of deploying LLMs in real-world human–robot dialogue. Our findings show that memory continuity, adaptive turn-taking, and culturally attuned voice design enhance user perceptions of trust, naturalness, and social presence. We also identify persistent limitations related to response latency, hallucinations, and expectation management. This work contributes design insights and architectural strategies for future LLM-integrated robots that aim to support meaningful, emotionally resonant companionship in socially assistive settings.
format Article
id doaj-art-aa33b517e73a4396b07f7f27979dbfc9
institution OA Journals
issn 2076-3417
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-aa33b517e73a4396b07f7f27979dbfc92025-08-20T02:23:08ZengMDPI AGApplied Sciences2076-34172025-06-011511637710.3390/app15116377Designing Social Robots with LLMs for Engaging Human InteractionMaria Pinto-Bernal0Matthijs Biondina1Tony Belpaeme2IDLab-Airo, Ghent University—imec, Technologiepark-Zwijnaarde 126, 9052 Ghent, BelgiumIDLab-Airo, Ghent University—imec, Technologiepark-Zwijnaarde 126, 9052 Ghent, BelgiumIDLab-Airo, Ghent University—imec, Technologiepark-Zwijnaarde 126, 9052 Ghent, BelgiumLarge Language Models (LLMs), particularly those enhanced through Reinforcement Learning from Human Feedback, such as ChatGPT, have opened up new possibilities for natural and open-ended spoken interaction in social robotics. However, these models are not inherently designed for embodied, multimodal contexts. This paper presents a user-centred approach to integrating an LLM into a humanoid robot, designed to engage in fluid, context-aware conversation with socially isolated older adults. We describe our system architecture, which combines real-time speech processing, layered memory summarisation, persona conditioning, and multilingual voice adaptation to support personalised, socially appropriate interactions. Through iterative development and evaluation, including in-home exploratory trials with older adults (<i>n</i> = 7) and a preliminary study with young adults (<i>n</i> = 43), we investigated the technical and experiential challenges of deploying LLMs in real-world human–robot dialogue. Our findings show that memory continuity, adaptive turn-taking, and culturally attuned voice design enhance user perceptions of trust, naturalness, and social presence. We also identify persistent limitations related to response latency, hallucinations, and expectation management. This work contributes design insights and architectural strategies for future LLM-integrated robots that aim to support meaningful, emotionally resonant companionship in socially assistive settings.https://www.mdpi.com/2076-3417/15/11/6377Large Language Modelsspoken language interactionsocial robotselderly users
spellingShingle Maria Pinto-Bernal
Matthijs Biondina
Tony Belpaeme
Designing Social Robots with LLMs for Engaging Human Interaction
Applied Sciences
Large Language Models
spoken language interaction
social robots
elderly users
title Designing Social Robots with LLMs for Engaging Human Interaction
title_full Designing Social Robots with LLMs for Engaging Human Interaction
title_fullStr Designing Social Robots with LLMs for Engaging Human Interaction
title_full_unstemmed Designing Social Robots with LLMs for Engaging Human Interaction
title_short Designing Social Robots with LLMs for Engaging Human Interaction
title_sort designing social robots with llms for engaging human interaction
topic Large Language Models
spoken language interaction
social robots
elderly users
url https://www.mdpi.com/2076-3417/15/11/6377
work_keys_str_mv AT mariapintobernal designingsocialrobotswithllmsforengaginghumaninteraction
AT matthijsbiondina designingsocialrobotswithllmsforengaginghumaninteraction
AT tonybelpaeme designingsocialrobotswithllmsforengaginghumaninteraction