.png)
Course description
This beginner-friendly, hands-on workshop introduces retrieval-augmented generation (RAG), a technique that enables AI systems to retrieve external information before generating responses. Participants will explore why a retrieval system can return information that appears relevant to a question but is unsuitable, misleading, or unsafe to use, particularly in high-stakes contexts such as health information.
Using practical examples and a case study inspired by maternal and neonatal health information retrieval, participants will examine how query wording, document selection, chunking strategies, and source quality can influence retrieval results. Through guided activities, they will learn to assess retrieved information beyond semantic similarity and consider how retrieval systems can be evaluated more responsibly.
The workshop is designed for beginners and participants with some experience in digital research, AI, or information retrieval. No prior experience with RAG is required.
Target audience
The workshop is intended for master's students, PhD candidates, researchers, librarians, digital scholarship practitioners, and other professionals interested in AI-assisted research and information retrieval.
Relevant fields include computer science, information science, digital humanities, library and information studies, health informatics, data science, and interdisciplinary research involving generative AI and responsible AI evaluation.
Learning outcomes
By the end of this workshop, participants will be able to:
- Explain the basic principles of retrieval-augmented generation (RAG) and the role of information retrieval in AI-generated responses.
- Distinguish semantic relevance from the reliability, suitability, and safety of retrieved information.
- Identify common retrieval problems arising from query ambiguity, document chunking, and source selection.
- Apply basic evaluation criteria to assess retrieved passages using practical examples.
- Discuss approaches to making retrieval-based AI systems more reliable and appropriate for their intended context.
Prerequisites
- Entry level: Beginner.
- No prior experience with retrieval-augmented generation, machine learning, or programming is required.
- A general interest in AI, digital research, information retrieval, or the responsible use of AI tools is sufficient. Basic familiarity with searching for and evaluating digital information would be helpful but is not essential.
Required materials
- Participants should bring a laptop with a modern web browser and have access to a reliable internet connection.
- The practical exercises may use Google Colab notebooks to explore retrieval and evaluate retrieved information without requiring local software installation.
- Participants may need a Google account to run the notebooks. The instructor will provide the notebooks, sample queries, and datasets.
- Any required tools and access arrangements will be communicated in advance.
Lunch
- There will be a complementary buffet-lunch at 12:00-13:00 from the Sverdrups spiseri (Georg Sverdrups Hus, 2nd floor).
- After you get your lunch, join us at Store møterom where we have reserved seats for #DSDays2027 participants.
Number of participants
20
Contact
If you have any questions regarding this course, or anything else about the DSDays 2027 event, please contact us per email: digitalscholarship@ub.uio.no