Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
Publications cover a wide array of topics such as formalizing RNA structure prediction using dependency parsing, analyzing changes in commercial machine translation systems over time, mitigating hallucination through concept erasure methods, defining 'languages' of neural architectures with context-free grammars, pushing symbolic modeling for document summarization, and exploring alternatives to probing by using matrix factorization for representation analysis. Books: A second edition of Bayesian Analysis in Natural Language Processing. Projects: Rainbow Parser, XSum abstractive summarization system, etc.
Research Experience
Position: Reader; Affiliation: Institute for Language, Cognition and Computation, School of Informatics, University of Edinburgh.
Background
Interests lie at the broad intersection of natural language processing and machine learning. Main areas of work include text generation, parsing in its various forms, and representation learning and analysis. Tools used range from large language models, neural networks, (multi)linear algebra to probabilistic grammars.
Miscellany
Teaching: Has taught courses on Natural Language Understanding, Generation, and Machine Translation, Accelerated Natural Language Processing, Foundations of Natural Language Processing, Processing Formal and Natural Languages, and Topics in Natural Language Processing.