Honours and Masters project

Ensuring construction quality and safety requires timely detection of defects, yet traditional manual inspections are slow, costly, and inconsistent. This research presents a computer vision-driven solution that automates defect detection using images captured by drones and site cameras. By applying advanced deep learning models, the system are expected to identify cracks, corrosion, and surface irregularities with high accuracy, even under challenging site conditions.

Vision-and-Language Navigation (VLN) has emerged as a promising paradigm for enabling robots to interpret natural language instructions and navigate complex environments. While most prior work focuses on single-agent navigation, this study extends VLN to multi-robot systems, addressing challenges of coordination, communication, and task allocation in dynamic settings.

Generative AI is increasingly being used to support feedback and reflection in education. However, most current tools rely on cloud-based models, raising concerns around privacy, data governance, and trust, especially when handling sensitive student teamwork data.

This project explores the use of locally deployed generative AI (GenAI) to support teamwork feedback in computing education contexts. The focus is not only on technical feasibility, but also on how students and educators perceive, trust, and interact with AI-generated feedback when it is processed locally.

Teamwork feedback tools are increasingly used in computing education to support reflection, coordination, and peer interaction. However, students often experience challenges in interpreting feedback and understanding how to act on it, particularly when interfaces are not designed with their needs in mind.

This project focuses on the user experience (UX) and interface design (UI) of teamwork feedback systems (e.g., Pulse or similar platforms). It explores how interface design influences students’ ability to interpret, trust, and act on feedback during team-based project work.

This interdisciplinary project explores how artificial intelligence (AI) can support the development of trauma-informed practice among pre-service teachers. Trauma exposure affects more than two-thirds of school students and has significant implications for their academic engagement, emotional wellbeing, behaviour, and social participation at school.

This project investigates and mitigates confirmation bias in retrieval-augmented generation (RAG) systems applied to scientific question answering in the health domain. RAG systems are increasingly used to answer clinical and biomedical questions by retrieving relevant publications and synthesising an answer with an LLM, but recent work shows that such pipelines can systematically prefer evidence that confirms the framing of a query while under-retrieving evidence that refutes it [1].

This project develops a focused prototype for LLM-assisted causal claim extraction and verification in mental health research. Clinical psychologists and psychiatrists increasingly rely on the rapidly growing biomedical literature to identify risk factors, evaluate treatments, and update practice, but the volume of new publications makes manual synthesis impossible.

Modern NLP applications increasingly process text carrying sensitive personal information, including clinical conversations, legal correspondence, customer support transcripts, and social media posts. Sharing such text with third-party models, annotators, or downstream pipelines remains constrained by data protection legislation (e.g., GDPR, the EU AI Act) and growing user expectations around transparency.

Modern large multimodal models (LMMs) and omni-modal models process not just text but vision, audio, and speech, opening new application surfaces and, with them, new safety risks. Established safety pipelines, including RLHF, safety classifiers such as Llama Guard, and red-teaming protocols, were largely developed for text-only models and translate poorly to the multimodal setting. Three gaps are now well documented.

Legal professionals routinely spend significant time on case analysis tasks that follow well-defined reasoning protocols, of which the Issue-Rule-Application-Conclusion (IRAC) method is the most widely taught and practiced framework across common-law jurisdictions.