I explore how these models can help robots perceive and interpret human behaviour through cues such as F-formations, gestures and gaze, supporting natural and adaptive interactions with visitors. My work involves designing and evaluating multimodal interaction techniques that connect visual perception, contextual understanding and robot behaviour. My broader research interests include human–AI interaction and human–computer interaction.
Zhikun Wu
PhD student
PhD student specialising in AI-based human–robot interaction.
Presentation
My research focuses on developing intelligent robot guides for museum environments using vision-language models and multimodal large language models.
Publications
2026
Teachers' Perceived Benefits and Risks of AI Across Fifty-Five Countries: An Audit of LLM Alignment and Steerability
L@S '26: Proceedings of the Thirteenth ACM Conference on Learning @ Scale
(Conference paper)
https://dx.doi.org/10.1145/3774398.3811616
Correction: Beyond the looking glass: multimodal LLM-based depth-sensing for spatial behavior modeling in media architecture (vol 8, 1746674, 2026)
Frontiers in Computer Science, Vol. 8, Article 1865390
(Article in journal)
https://dx.doi.org/10.3389/fcomp.2026.1865390
Beyond the looking glass: multimodal LLM-based depth-sensing for spatial behavior modeling in media architecture
Frontiers in Computer Science, Vol. 8, Article 1746674
(Article in journal)
https://dx.doi.org/10.3389/fcomp.2026.1746674
"Same Voice, Different Language": An Exploration of Voice-Cloned Translation to Support Non-Native Speakers in Online Meetings
31st International Conference on Intelligent User Interfaces, Paphos, Cyprus, March 23-26
(Conference paper)
https://dx.doi.org/10.1145/3742413.3789074
