Final Project Report and Oral Examination

Information retrieval is a field concerned with the structure, analysis, organisation, storage, searching, and retrieval of information. We will discuss advanced aspects of these topics, focusing on recent state-of-the-art developments that have been made, especially concerning neural models. Particular attention will be paid to different classes of neural rankers and how they compare in terms of their effectiveness at ranking and their efficiency. The module will be evaluated in terms of a small research project and an oral exam, which we will prepare for throughout the module with in-class discussion sessions.

Students will deeply understand the theory, design, and implementation of neural transformer ranking models, how to develop and measure effective and efficient search engines, including retrieval-augmented generation (RAG), and how to apply their learnings to domain-specific scenarios.

Teilnahmevoraussetzungen:
Vorheriges Absolvieren des Moduls "Modern Search Engines" (nicht verpflichtend, aber empfohlen)

----

Prerequisites:
Previous completion of the module "Modern Search Engines" (not obligatory, but recommended).