Generative Artificial Intelligence (GenAI) has not only captured the imagination of the research community but has also inspired government, industry and the general public. One of the key developments in GenAI has been Large Language Models (LLMs). While the technical implementation of LLMs is reasonably well understood, their application and how they can be best used is not well understood. At present it is fair to say that everyone is trying to understand how to use and deploy GenAI, and LLMs in particular, in daily use scenarios. One approach has been prompt engineering; how to best query LLMs (like ChatGPT, etc.) in order to obtain the best response or most accurate/detailed response.
This project will undertake a comprehensive study and analysis of existing prompt engineering techniques. Moreover, it will develop guidelines for the use of LLMs. The project might also consider classifying different types of problem domains and identifying the most appropriate prompt engineering techniques for these domains. In summary, the main aim of this project is to put more rigour around the use of GenAI; in particular, LLMs.
School
Computer Science and Engineering
Research Area
Artificial Intelligence | Deep learning | Large language models
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
You will work closely with the academic supervisors and other PhD students in the group.
A comprehensive summary and analysis of existing prompt engineering techniques for LLMs. The development of a principled and rigorous approach to using GenAI, in particular LLMs, when applied to difficult problems.
