about me
Hi. I’m a Ph.D. candidate in the Institute for Language, Cognition, and Computation at the University of Edinburgh. I also did my B.Sc. in Computer Science at Brown University. I research techniques in natural language inference (NLI), especially to do with tasks like question answering, and data in the form of structured knowledge (Knowledge Graphs, Entailment Graphs) and unstructured, open-domain natural language relations. I'm also interested in Language Model capabilities in language understanding, as well as representation learning and latent structure in Language Model embedding space. In the past I've made some creative things, like games, apps, and generative art. Check out a sample of my publications and projects below.
selected publications
Nick McKenna*, Tianyi Li*, Liang Cheng, Mohammad Javad Hosseini, Mark Johnson, and Mark Steedman. Sources of LLM Hallucination by Large Language Models on Inference Tasks. ArXiv, 2023. Paper Link. Nick McKenna and Mark Steedman. Smoothing Entailment Graphs with Language Models. ArXiv, 2022. Paper Link. Nick McKenna, Liane Guillou, Mohammad Javad Hosseini, Sander Bijl de Vroe, Mark Johnson, and Mark Steedman. Multivalent Entailment Graphs for Question Answering. EMNLP, 2021. Paper Link.
creative projects
Reach me by email or find me on Twitter and LinkedIn. I'd love to hear from you!
© 2022 Nick McKenna