Human-Computer Interaction
Designing and evaluating interactive systems that support cognition, learning, decision making, and user experience.
University of Auckland PhD Student
Lefan Lai
My name is Lefan Lai (赖乐凡), and I am currently pursuing a PhD at the University of Auckland. Thank you for visiting my personal website.
01 / Human
The person is not a user label.
My HCI work begins with people as learners, searchers, collaborators, and spatial thinkers.
Spatial skills and STEM learning
Studying how interactive tools can help people practise spatial reasoning and develop skills connected to STEM learning.
User studies as research instruments
Using empirical studies to understand attention, memory, perception, task demands, and human skill development.
02 / AI
The model is not the whole intelligence.
I also work on technical AI questions, especially reliable and human-centred methods for LLMs and multimodal systems.
Large language models
Investigating LLMs and multimodal LLMs as systems that reason with language, vision, uncertainty, and context.
Hallucination and uncertainty
Exploring ways to reduce object hallucination and improve reliability through uncertainty-guided visual refinement.
Training-free refinement
Studying efficient methods that improve model behaviour without requiring expensive retraining pipelines.
03 / Space
Space changes what we can notice.
XR and mixed reality let research questions become embodied: objects, depth, memory, motion, attention, and skill can all be studied together.
Extended reality
Exploring how virtual and physical environments shape perception, attention, interaction, and spatial behaviour.
Mixed reality visual search
Understanding how people search for virtual information in complex environments and how spatial regularities influence memory.
Spatial skills training
Designing adaptive interactive tools and games that support spatial reasoning and human skill development.
News
Research notes
2025 – 2026
Issue 01
Category
Paper Status
Showing 9 updates.
PaperSubmitted
I submitted a paper to ICLR 2027 as the second author.
I contributed to the submission as its second author.
PaperSubmitted
I submitted two papers to CHI as the second author.
I contributed to both submissions as their second author.
PaperAccepted
VIGOR was accepted to Findings of EMNLP 2026.
Our paper Looking Closer When Unsure: VIGOR, A Training-Free Framework for Mitigating Object Hallucination in Multimodal Large Language Models was accepted to Findings of EMNLP 2026.
PaperSubmitted
I submitted my first lead-authored AI paper to AAAI 2027.
This marked a new step in my technical AI research and my first time leading an AI paper.
PaperSubmitted
VIGOR was resubmitted to EMNLP 2026.
We submitted a revised version of our multimodal AI paper to EMNLP 2026.
Social
I attended CHI 2026 in Barcelona.
I travelled to Barcelona for the CHI conference and presented Searching Through Complex Worlds: Visual Search and Spatial Regularity Memory in Mixed Reality.
PaperSubmitted
My first AI paper was submitted to EMNLP 2026.
We submitted Looking Closer When Unsure: VIGOR, the first AI paper I contributed to, to EMNLP 2026.
PaperAccepted
My first paper was accepted at CHI 2026.
Searching Through Complex Worlds: Visual Search and Spatial Regularity Memory in Mixed Reality became my first accepted publication.
PaperSubmitted
I submitted my first lead-authored paper to CHI 2026.
The submission grew from my master's research on visual search and spatial regularity memory in mixed reality.
Publications
Showing 2 papers.
Searching Through Complex Worlds: Visual Search and Spatial Regularity Memory in Mixed Reality
ACM Conference on Human Factors in Computing Systems
Looking Closer When Unsure: VIGOR, A Training-Free Framework for Mitigating Object Hallucination in Multimodal Large Language Models
Conference on Empirical Methods in Natural Language Processing
About
I am currently a PhD student at the University of Auckland, New Zealand. My PhD scholarship is funded through a Smart Ideas project supported by MBIE. I am jointly supervised by Associate Professor Burkhard Wuensche and Professor Mark Billinghurst, with each supervisor contributing equally to my PhD supervision. I am affiliated with both the Graphics Lab and the Empathic Computing Lab.
Before starting my PhD, I completed a Master of Computer Science at the University of Sydney, where I worked in the AID-Lab under the supervision of Dr Brandon Syiem and Dr Zhanna Sarsenbayeva.
I received my Bachelor's degree in Software Engineering from Nanchang University. My research interests include Human-Computer Interaction, Artificial Intelligence, Large Language Models, Extended Reality and Mixed Reality, multimodal AI, spatial skills training, visual search, and human-centred intelligent systems.
PhD in Human-Computer Interaction
The University of Auckland, New Zealand.
Master of Computer Science
The University of Sydney, Australia.
WAM 85.75 High Distinction
Bachelor of Software Engineering
Nanchang University, China
Open conversation
I am always happy to discuss any question, explore unfamiliar territory, and follow ideas wherever curiosity leads.
Lefan Lai. 赖乐凡, Auckland, New Zealand. A question about people. A problem for AI. An idea worth discussing.
