I am a third-year PhD student at the Social Robotics Lab, Uppsala University. Recently, my research interests mainly concern Large Language Models in topics such as Knowledge Manipulation, Personality Expression, Multi-Modal Generation in Social Robots and all the ethical implication involved. I'm also part of PREVENT, an Uppsala-based project for cardio-metabolic diseases prevention in disadvantaged populations.
I obtained my Master's in Machine Learning at KTH, Sweden in 2022. My master's thesis was in the field of social robotics where I explored the incorporation of emotions and personality into a conversational social agent (see links below).
I obtained my bachelor's degree in Computer Science with honours at the University of Pisa, Italy. My bachelor's thesis investigated the use of whole positive numbers in the traditional switching theory with applications e.g. in the field of data mining.
At Uppsala University, I've been a teaching assistant in the course "Intelligent Interactive Systems" since 08/2022, Social Robotics since 01/2024 and "Large Language Models and Societal Consequences" since its first iteration.
At KTH, I've been a teaching assistant in the master-level courses: "Artificial Intelligence" from 12/2020 to 06/2022, "Deep learning in Data Science" from 03/2022 to 06/2022 and "Search Engines and Information Retrieval Systems" from 01/2022 to 06/2022.
I'm open to master's thesis and internships on Large Language Models in selected topic such as: knowledge augmentation and editing in language models, multi-modal generation for social robots, moral and ethical reasoning.
Strong knowledge of PyTorch and Language Models is required. Willingness to pursue a publication is an advantage.
For more information: alessio.galatolo@it.uu.se
May '25: My paper "Visualising Policy-Reward Interplay to Inform Zeroth-Order Preference Optimisation of Large Language Models" was accepted at ACL 2025 Findings.
Apr '25: My paper "Simultaneous text and gesture generation for social robots with small language models" was accepted for Frontiers in Robotics and AI.
April '24: Arrived third at the ImpactAI hackathon. Our project was developed for and in collaboration with the World Childhood Foundation.
Nov '24: Wrote an article for SVEP (pp. 8)
Sept '24: Presented my research at UDC's lunch seminar
June '24: A new pre-print is available for my paper: "DIEKAE: Difference Injection for Efficient Knowledge Augmentation and Editing of Large Language Models"
June '24: A new pre-print (now accepted to JAIR!) is available for the paper in collaboration with the University of Bologna: "Promoting Fairness and Diversity in Speech Datasets for Mental Health and Neurological Disorders Research"
Nov '23: Gave a presentation at Arkus AI, marking the beginning of our collaboration
Oct '23: My (student) paper has been accepted at AAAI24: "Memory-Augmenting Decoder-Only Language Models through Encoders".
Aug '23: Presented at RO-MAN 2023 the paper “Personality-Adapted Language Generation for Social Robots”. Learn more here
May '23: My first-author paper “Personality-Adapted Language Generation for Social Robots” was accepted at RO-MAN 2023. Learn more here
April '23: Presented a robot demo to stakeholders at Uppsala for the PREVENT project
March '23: Joined HRI2023 as a student volunteer
February '23: Built, in collaboration with KTH’s social robotics, 3 awesome Blossom robots