Cohort 2019

Georgia Carter

Thesis Defended: Effects of Context on Semantic Representations and Mechanisms in Humans and Language Models

Supervisors: Paul Hoffman, Frank Keller

Jie Chi

Thesis Defended: Understanding and Modeling Code-Switching: Metrics, Triggers, and Applications in Multilingual NLP

Supervisors: Peter Bell, Catherine Lai

Henry Conklin

Thesis Defended: Information Structure in Mappings: An Approach to Learning, Representation, and Generalisation

Supervisors: Kenny Smith, Ivan Titov

Tom Hosking

Thesis Defended: Learning Weakly Structured Representations for Text-to-Text Generation

Supervisors: Mirella Lapata, Hao Tang

Parag Jain

Thesis Defended: Context Representations for Conversational Natural Language Understanding

Supervisors: Mirella Lapata, Ivan Titov

Faheem Kirefu

Research interests: Machine translation (in particular, low resource techniques) and natural language understanding

Research Topic: Parameter Efficient Fine Tuning

Supervisors: Lexi Birch, Barry Haddow

Nicole Meng-Schneider

Research Interests: Usable security, looking at the privacy implications of voice-controlled interfaces, especially smart speakers, and the social impact of such systems

Research Topic: OF SMART SPEAKER IN MULTI-USER SPACES 

Supervisors: Nadin Kokciyan, Maria Wolters

Rimvydas Rubavicius

Thesis Defended: Processing Embodied Conversation for Interactive Task Learning

Supervisors: Alex Lascarides, Ram Ramamoorthy

Socials: LinkedIn Google Scholar Personal Website

Rohit Saxena

Research interests: Efficient long-context models and multimodal reasoning

Research Topic: Abstractive Summarization of Long Narratives through Content Selection and Model Scaling

Supervisors: Frank Keller, Hao Tang

Socials: LinkedIn Google Scholar Personal Website

Emelie Van De Vreken 

Research Interests: Machine learning for speech synthesis and prosody modelling

Research Topic: Voice Puppetry for Synthesis Control

Supervisors: Korin Richmond, Catherine Lai

Dan Wells

Research Interests: Investigating speech synthesis for low-resource languages through multilingual modelling, transfer learning and model adaptation in end-to-end neural network systems.

Research Topic: On Low-Resource Text-to-Speech Synthesis

Supervisors: Korin Richmond, Hao Tang

Irene Winther

Research Interests: Mechanisms of lexical activation, the organisation of the mental lexicon, semantic context, prediction, bi-/multilingualism and computational modelling

Research Topic: Non-native Lexical Access: Insights from Computational and Behavioural Approaches

Supervisors: Martin Pickering, Yevgen Matusevych