Large language models, such as GPT-3 and its variants, have been making significant advancements in the field of artificial intelligence. One of the challenges these models face is adapting to new tasks with limited training data, known as few-shot learning.

The aim of this thesis is to investigate the current state of few-shot learning with large language models and evaluate the effectiveness of structured prompting and prompt-tuning methods in this context.

• Preliminary knowledge on machine learning, data science
• Python programming
• Good understanding of the mathematical foundations of optimization and linear algebra
   

If you are interested in a Bachelor thesis, please write a meaningful email that addresses your previous experience, interests, and strengths.

We investigate structured prompting and prompt-tuning methods on a dataset of patient interviews. The results will be analyzed and compared with existing methods to determine the effectiveness of the proposed approach.

1. https://arxiv.org/abs/2211.05100
2. https://arxiv.org/pdf/2101.00190.pdf
3. https://arxiv.org/pdf/2210.02441v3.pdf