2023 - 2024 Course Highlights DissertationsMachine Learning:Controlled Sampling in Diffusion ModelsEfficient and Unbiased Sampling of Molecular Energy Functions via Consistency ModelsEnergy-Based Diffusion Neural Sampler for Boltzmann DensitiesEvaluating Backdoor Defense Techniques for Large Language ModelsEvaluating Backdoor UnlearningGenerative Models for Synthesizable LipidsLearning to Forget with Diffusion HypernetworksMitigating and Assessing Bias and Fairness in Large Language Model-Generated Synthetic Tabular DataNonlinear, Scalable Sequential Component Analysis{On, Off}-The-Grid Data Modelling Using Transformer Neural ProcessesOptimal Path Flow for Multimodal GenerationProcess-Centric Gaussian Processes Speech and Language Processing:Automated Situated Task Guidance with Foundation ModelsDirect Preference Optimisation for Natural Language Generation with Minimum Bayes Risk DecodingReranking Multi-Modal Retrievers for Knowledge-Based Visual Question AnsweringRetrofitting Language Models with Dynamic Tokenisation Second-Order Optimisation and Imbalanced Class Distribution in Emotional Analysis Computer Vision and Robotics:A Vision-Language Model as Embodied Manufacturing AgentIdentifying Looted ArtefactsLanGWM: Language Grounded World ModelModelling Additive Manufacturing Processes via Graph-Conditioned Diffusion ModelsReservoir Computing on Mobile Swarm Dynamical Systems Human-Computer Interaction:Supervisory Control of a Multi-Arm Telerobotic System Supported by Mixed Reality
