Photo of Greg Greg Durrett
Associate Professor, UT Austin Computer Science

CV / TAUR lab / UT NLP

Research Group and Publications

My students, collaborators, publications, and other information about my research group can be found on the TAUR lab website.

Google Scholar / Semantic Scholar

My research is primarily in the field of Natural Language Processing. My group focuses on improving techniques for accessing and reasoning about knowledge in text. Large language models like GPT-4 have dramatically advanced these frontiers; currently we are looking at where these systems succeed and fail and how to further enhance their capabilities, particularly by building modular NLP systems that use LLMs as primitives. Some examples of recent projects include:

  • Natural language deduction: training systems to make grounded deductive inferences from provided evidence
  • Learning models to assess fine-grained factuality of summarization systems
  • Automatic fact-checking with LLMs
  • Advancing fundamentals of in-context learning and inference for generation with LLMs


CS388: Natural Language Process (Online MSCS/MSDS version): video lectures, readings, and assignments for the online masters offering of these NLP courses.

NLP Module for high schools: videos and hands-on assignments introducing n-gram language models and pre-trained Transformers.

Undergraduate and graduate teaching:

Read about some of the course materials we use in our TeachingNLP workshop paper:

Contemporary NLP Modeling in Six Comprehensive Programming Assignments
Greg Durrett, Jifan Chen, Shrey Desai, Tanya Goyal, Lucas Kabela, Yasumasa Onoe, Jiacheng Xu. Proceedings of the Fifth Workshop on Teaching NLP.


You can find videos of courses, talks, and more on my YouTube channel. Some of these are silly videos.

When I'm not working, I'm often struggling with various physical pursuits (running, cycling, climbing, and hiking). My half marathon PR is 1:55:26 (set 2/19/2023).