![Thomas Greatrix BSc](https://profiles-cardiff.imgix.net/attachments/4845?w=200&h=200&auto=format&crop=faces&fit=crop&q=90")
Publication
2024
- Greatrix, T., Whitaker, R., Turner, L. and Colombo, W. 2024. Can large language models create new knowledge for spatial reasoning tasks?. [Online]. arXiv: Cornell University. Available at: https://arxiv.org/abs/2405.14379
Websites
- Greatrix, T., Whitaker, R., Turner, L. and Colombo, W. 2024. Can large language models create new knowledge for spatial reasoning tasks?. [Online]. arXiv: Cornell University. Available at: https://arxiv.org/abs/2405.14379
Research
I research whether AI can consistently produce interesting artifacts: A baker may be interested in chocolate cake recipes, a mathematician might be interested in proofs, or a artist might be interested in new art. In all these cases, AI can be used to create items that these people find interesting. Unfortunately, if they were to repeatedly ask the AI to give them interesting stuff repeatedly, they would often start to get "samey" results. My research tries to extend the amount of interesting items that can be produced, with the aim of being able to produce a range of interesting items in perpetuity.
Contact Details
Research themes
Specialisms
- Artificial intelligence
- creative methods