I lead the Multimodal Learning Lab, a joint research group of the University of Aberdeen and the Nepal Applied Mathematics and Informatics Institute for Research (NAAMII), Kathmandu. We develop reliable multimodal learning for clinical and industrial environments where labelled data, access to modalities and computational resources are constrained.
The group brings together clinical and computer science PhD researchers, an MSc by Research student and research assistants at NAAMII. Projects connect surgical perception, hallucination detection, federated learning, edge AI and low-resource language models. The researchers and their project areas are listed below.
Our training pathway links participation in research projects with publication and progression to graduate study and research careers. Former members and mentees have progressed to institutions including MILA, Dartmouth, Stanford, Inria and universities across Europe and North America, as well as research roles in industry. Their individual pathways are recorded below.
We build datasets and benchmarks with clinical collaborators and treat these resources as research contributions. The group’s work connects methodological development with evaluation in the environments where reliable predictions matter.
Senior Lecturer (Associate Professor), University of Aberdeen, UK
Co-founder and Adjunct Research Scientist, NAAMII, Nepal
Chief AI Scientist, Fogsphere (Redev AI Ltd), UK
Jardin Ruari
Clinical PhD Student
Assessing AI Algorithms for Capsule Endoscopy
Krit Duangprom
CS PhD Student
Surgical Tool and Hand Pose Estimation
Aminu Lawal
CS PhD Student
Hallucination Detection in VLMs
Love Panta
CS MSc by Research Student
Surgical Vision World Models
Bibek Niroula
Multimodal Federated Learning
Saurav Bastola
Synthetic Data Generation
Niyoj Oli
Hallucination Detection in VLMs
Aavash Chhetri
Multimodal Federated Learning
Dipan Bartaula
Edge AI
Anurag GC
Spatial Understanding in VLMs
Sagar Giri
Nepali LLMs
Deepak Thapa
Reasoning Models