https://dasya.itu.dk/semester/spring-2026/ Recent content in Spring 2026 on DASYA Hugo -- gohugo.io en-us © DASYA, IT University of Copenhagen Fri, 21 Nov 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/osk-qa-multimedia/ Fri, 21 Nov 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/osk-qa-multimedia/ Constrastive learning models have made it easier to find relevant content from multimedia collections through descriptive text queries, reducing the interactivity needed to solve simple tasks. However, for complex tasks not only pertaining visual elements, or tasks focused on answering questions involving the contents of a singular or multiple media items (videos/images) still requires a fair amount of interactivity. Retrieve results from query Inspect content (summaries, watching, browsing) Refine the query / add more context / different search strategy Repeat until answer to the question or relevant content has been discovered RAG (Retrieval-Augmented-Generation) approaches utilize LLMs (Large-Language Models) to answer questions from a set of documents. https://dasya.itu.dk/for-students/proposals/archive/mh-llm-vs-rng/ Thu, 20 Nov 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/mh-llm-vs-rng/ BSc Thesis / MSc Thesis. The goal of this project is to explore the feasibility and impact of replacing traditional random number generators (RNG) in games with large language models (LLMs). For example, in a slot machine game, the outcome is typically determined by a random number generated each time the “spin” button is pressed. In this project, students will investigate using an LLM to generate these numbers instead. Research questions include: https://dasya.itu.dk/for-students/proposals/archive/mh-charmaps/ Wed, 02 Jul 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/mh-charmaps/ MSc Thesis. Character maps in database systems are specialized data structures optimized for efficient string handling. Similar to Bloom filters, they allow quick checks for the presence of characters or substrings without full string comparisons. This makes them useful for accelerating string-related queries and improving overall search performance in databases. The goal of this project is to delve deeply into character maps: What strategies for building character maps exist, and how do they compare to Bloom filters and prefix trees (aka tries)? https://dasya.itu.dk/for-students/proposals/archive/mh-js-framework-udf/ Wed, 02 Jul 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/mh-js-framework-udf/ BSc Thesis. The idea of this project is to write a game framework in JavaScript (e.g., a physics engine or state machine that keeps track of the game state) that can be re-run deterministically using databases. That is, in addition to being a standalone framework, the JavaScript framework must run as a user-defined function (UDF) in a relational database where it can be executed on thousands of existing user inputs. The research questions include whether such a framework is feasible to build, how fast it executes within a database system as a UDF, and how it scales when executed against thousands or more existing game states or user input sessions. https://dasya.itu.dk/for-students/proposals/archive/mh-minmax-duckdb/ Wed, 02 Jul 2025 00:00:00 +0000 https://dasya.itu.dk/for-students/proposals/archive/mh-minmax-duckdb/ MSc Thesis. DuckDB stores data in a database file. The database file is split into partitions, and for each partition, DuckDB keeps statistics in the form of min-max summaries. Similar min-max summaries exist in Parquet files and many other database systems, such as Snowflake. Check out Lecture 8 of last semester’s Introduction to Database Systems course for a summary of min-max summaries (I can send the slide deck if needed).