Summary

Large Language Models (LLMs) are increasingly becoming important tools in everyday applications. One of their most interesting capabilities is in-context learning, which enables them to perform new tasks by providing examples or instructions within the input, without requiring explicit model retraining.

At the core of modern LLM development are foundation models, which have demonstrated remarkable success in natural language processing. More recently, this progress has expanded beyond text, with increasing research exploring the use of foundation models and LLMs for tabular data.

This MSc thesis will investigate the capabilities of LLMs for learning from and reasoning about tabular data, with a particular focus on in-context learning and its potential applications in data analysis and prediction.