Advances in Database Research
News
- The kickoff meeting will take place on Thursday, Oct 9, 2025, 14:00 in seminar room SR T06. Attendance is mandatory.
- The overarching topic of this seminar will be vector search.
Schedule
Location: Seminar room T06
About
Questions and discussions
For questions and discussions (also among students) regarding course specific topics please use the Slack channel #adr-se (Workspace dbteaching.slack.com).
Slack registration: Students register with their university email here: https://dbteaching.slack.com/signup
Procedure and Evaluation
The seminar consists of presentations given bei the students and discussions on the presentations.
Presentations: The students choose a current research paper (from a given list), which they prepare and present during the seminar. The presentation should be very detailed and didactically well prepared. Students should not just recite the explanations in the research paper, but acquire the necessary background knowledge and build their own understanding. Students should be able to respond to detailed questions and create and solve examples on their own.
Rehearsal: Each presentation will be preceded by a rehearsal, which takes place two weeks (or two seminar sessions) before the actual presentation.
The rehearsal will be attended by the presenting student and two other participants of the seminar, who act as peer mentors. Each student serves as a peer mentor for two other presentations. Peer mentors are expected to read and understand the corresponding research papers in advance, provide constructive feedback during the rehearsal, and help the presenter improve the content and delivery.
After each rehearsal, the peer mentors write a short report about the rehearsal and the presented paper. During the actual seminar presentation, peer mentors are also encouraged to contribute to the discussion and assist in answering questions from the audience.
Discussion: students listen to presentations by fellow students and young researchers and engage with the content in discussions.
The quality of the presentation, the rehearsal report, the participation in the discussions, and the quality of the contributions to the discussion will be evaluated.
Attendance: Please note that attendance at the presentations is mandatory. The required attendance rate is 80% (including the presentations and the kickoff meeting).
Attendance at the mentoring sessions is also mandatory for the sessions that you are assigned to mentor. These mentoring sessions do not count toward the 80% attendance requirement.
If you miss one of your mentoring sessions due to events beyond your control (e.g., illness documented by a medical certificate), you will be required to mentor an additional paper to make up for it.
Research Papers
The overarching topic of this seminar will be vector search.
Vector search enables databases to retrieve information based on meaning and similarity rather than exact matches. By transforming text, images, audio, or even code into numeric vectors, it allows queries like “how close in meaning is this item?” rather than simply “does this keyword appear?” This approach powers semantic search, recommendation systems, and many AI applications. The main challenges lie in scaling to large vector databases while maintaining high throughput and low latency, and in seamlessly combining vector search with exact predicate filtering.
The list of research papers will be announced soon on our Slack channel #adr-se.
Please select one paper to present and two papers to mentor by October 23, 2025. A registration list will be available on our Slack channel.