Natural Language Understanding

General Resources

  • Leeds NLU Group research literature repository. This is a large collection of publications relevant to natural language understanding and the Winograd Schema Challenge.
  • OpenAI provides very powerful and interesting tools for investigating language, mainly using generative models based on GPT3.
  • BERT Explained: What it is and how does it work? | Towards Data Science
  • HuggingFace BERT interface
  • ConceptNet is a large multi-lingual semantic network. You can look up a word and see links to other related words and phrases.
  • Linguistics for the Age of AI, an open-access textbook about AI-based linguistics.
  • FRED is a website about Machine Reading for the Semantic Web, in other words how to extract semantic information from natural language. It includes an interface where you can input text and get a semantic representation in the form of a tree.
  • Connective-Lex.info is an online resource that hosts detailed structured information concerning connective words and their meanings.
  • SENNA is software that provides syntactic and semantic parsing of natural language including semantic role labelling.

The Winograd Schema Challenge

The Winograd Schema Challenge was proposed by Hector Levesque of the University of Toronto as a test for artificial intelligence, similar to the Turing Turing test. The test is based on a problem of natural language understanding that was presented in an example given by the pioneering AI researcher and computational linguist Terry Winograd. Levesque suggested that this kind of problem can form the basis of a test for human-level understanding of natural language text that is more objective than the Turing Test.

Our Publications

  • Hong, Bennett, Clymo, Alvarez (2022) KARaML: Integrating Knowledge-Based and Machine Learning Approaches to Solve the Winograd Schema Challenge, AAAI-MAKE Spring Symposium. [Download]

  • Bennett (2021) Semantic Analysis of Winograd Schema No. 1, Proceedings of the 12th International Conference on Formal Ontology in Information System (FOIS 2021). [Download]

  • Hong and Bennett (2021) Tackling Domain-Specific Winograd Schemas with Knowledge-Based Reasoning and Machine Learning. [Download]

  • Hong and Bennett (2020) Tackling Domain-Specific Winograd Schemas with Knowledge-Based Reasoning and Machine Learning (ArXiv version, superceeded by above LDK-2021 paper). [Download]

  • Suk Joon Hong (2020) Tackling Domain-Specific Winograd Schemas with Knowledge-Based Reasoning and Machine Learning, MSc Dissertation. [Download]