ISTD PhD Oral Defence Seminar by Mark He Huang – Modern AI systems often struggle in physical environments where observations are incomplete, noisy and constantly changing. This seminar presents a feedback-guided representation learning framework that helps AI systems revise what they know, maintain useful representations over time and generate physically plausible outcomes. The work spans machine unlearning, object-centric 3D visual memory and simulator-guided video generation, with the broader aim of enabling more reliable reasoning in the dynamic physical world.

ISTD PhD Oral Defence Seminar by Duo Peng – In this paper, we investigate Diffusion-Based Domain Adaptation, leveraging emerging
diffusion models to address domain adaptation tasks. The motivation behind our research stems from the powerful distribution transformation capabilities of diffusion models, which we aim to harness to help AI models adapt to new data distributions.

Congratulations to Associate Professor Liu Xiaogang’s PhD Student in Winning the Best Oral Presentation Award

ISTD PhD Oral Defence Seminar by Wang Tianduo – This thesis studies how large language models (LLMs) generalise under distribution shift.

ISTD PhD Oral Defence Seminar by Hong Pengfei – This thesis addresses critical questions surrounding the evaluation and enhancement of reasoning robustness, generalisability, and comprehensiveness in modern language models, particularly under realistic conditions involving noise, ambiguity, domain shifts, and multimodal inputs.

ISTD PhD Oral Defence Seminar by Zhu Lanyun – Semantic segmentation is a fundamental and important task in computer vision, which aims to classify each pixel in an image. The rapid development of deep learning has significantly advanced semantic segmentation and improved the accuracy, promoting its application in fields with high accuracy requirements for pixel-level prediction, such as autonomous driving and medical diagnosis. Current works for semantic segmentation are typically based on a standard setup that all data is accessible beforehand and can be learned simultaneously.

ISTD PhD Oral Defence Seminar by Perry Lam – Neural networks are known to be over-parametrised and sparse models have been shown to perform as well as dense models over a range of image and language processing tasks. However, while compact representations and model compression methods have been applied to speech tasks, sparsification techniques have rarely been used on text-to-speech (TTS) models. We seek to characterise the impact of selected sparse techniques on the performance and model complexity.

ISTD PhD Oral Defence Seminar by Chin Wai Kit Daniel – Social media and social networking platforms have greatly connected people worldwide and democratised information creation and propagation by facilitating seamless and almost instantaneous information sharing between people and communities.

ISTD PhD Oral Defence Seminar by Chia Yew Ken – This thesis investigates the crucial role of structured information in natural language processing and artificial intelligence, with a focus on its extraction, utilisation, and extension to multimodal reasoning.

Congratulations to PhD student Hee Ming Shan for obtaining SDSC Dissertion Research Fellowship 2023