Bio
Giulio Antonio Abbo is a PhD student in Computer Science and Engineering at Ghent University and researcher at IDLab (imec). His work focuses on human-robot interaction and dialogue systems, with an interest in contextual understanding, cognitive architectures, and value alignment. His research spans AI, robotics, and human-centred design, aiming to create more meaningful and ethical interactions between humans and machines.
Topics of interest
- Human-Robot Interaction
- Social Robotics
- Conversational Agents
- Human Values in AI and Robotics
- Value-Aware AI
Research
Here you will find an overview of my research.
From the latest work on the VALAWAI project – where I try to understand what awareness means for social robots and identify the values relevant in human-robot interactions – through my exploration of language models as tools for working with values, to my early research on conversational troubleshooting, each research line below is a collection of related publications.
RESEARCH LINE
VALAWAI
Most of my work as a PhD student revolves around the VALAWAI project, funded by the European Union’s Innovation Council.
It focuses on developing VALue-AWare Artificial Intelligence systems that integrate human values into their functioning.
Its goal is to enhance the ethical alignment and trustworthiness of AI applications by ensuring that they operate in ways that reflect the preferences of the people involved.
RESEARCH LINE
Embodied Awareness
When I started working on the VALAWAI project, I had no idea what the people writing the proposal meant by “awareness”.
In this research line, I explore what machine awareness is and how we can improve contextual awareness, especially in social robots.
Publications
A list of my publications, divided by type and sorted by date. Most preprints will be included after publication.
PUBLICATION
Multi-Party Open-Ended Conversation With a Social Robot
In this work we explore wether an LLM can guide a multi-party spoken interaction. To track speakers, in our setting face recognition performed best; voice recognition was unreliable due to frequent short utterances and overlapping speech. The response and its addressee were generated by the LLM. Voice direction of arrival and gaze tracking were used to determine when the Furhat robot should speak and when the participants were talking to each other. The way we handled the text generation and the robot’s speech gave some speed improvements and allowed participants to interrupt the robot’s speech if needed. First online 15 Jan 2025.
PUBLICATION
On-the-Fly Expressions for Furhat
In this paper, we present Expressive Furhat, a system that generates real-time facial expressions for the Furhat robot using large language models based on the dialogue context.
Posts
Interesting finds, lessons learned, and some tutorials, sorted by date and sometimes cross-posted on dev.to.
BLOG POST
Value-Aware Social Robots at the 2026 EIC Summit
Would you want your robot assistant to never lie? Should it always tell the truth, even if it might ruin your surprise birthday party? Should it report a stolen candy or a secret cigarette? How should it handle complex situations where human values conflict? These are the questions our work on value-aware social robots seeks to address.
BLOG POST
Three Tips for Your Next (Software) Demo
Implementing something is always only half of the work; the rest is, well… showtime! An exciting demo can make the difference between inspiring the world with our creations and not even being noticed. Here are three tips we learned from participating in the Flanders Technology & Innovation festival in Antwerp, in March 2024.