Kategorie: ‘Theses’
Immersive Vocabulary Learning with Large Language Models
The emergence of large language models (LLMs), along with recent advances in mixed reality (MR) and virtual reality (VR), enable new opportunities for applying virtual agents in education. These simulated humans can imitate real-life situations and interactions with native speakers, which leads to an immersive and engaging learning experience. Especially in VR, interactions can be simulated with the agents. This way, languages can be learned and practiced in realistic scenarios. The LLM has the potential to overcome the limitations of learning applications with pre-scripted scenarios as the LLM can react dynamically to the learner’s actions and can lead to personalized interactions.
Dynamic Co-Speech Gesture Generation for Tutoring Agents
Large Language Models (LLM) can be applied to transform a natural language (NL)-based text input query into a NL-based text answer. A common use case are personal assistants, e.g., for learning activities. In such teaching contexts they can process knowledge recorded in plain text documents, create summarizations, or teach knowledge according to a curriculum. However, interfaces for LLMs are currently text-based chats. This can be enhanced by showing body language, e.g., gestures which support the conveyed content. With the help of desktop-based virtual agents, the chat interface can be turned into a video call where the LLM is personified by an agent which is able to respond with gestures in addition to the output text.
DECODE: Data Explainability Concepts and Ontological Design Evaluation
The aim of this thesis is to evaluate and extend a developing ontology of explainable data principles, an ongoing work aimed at establishing a structured framework for Data Explainability in AI systems. The current version of this ontology is in its early stages, primarily focused on defining key principles of Data Explainability and exploring their role in enhancing trust and transparency in AI. This thesis will build on this preliminary work by refining the ontology’s structure, proposing new principles or dimensions where needed, and assessing its applicability across various AI domains to make it a robust foundation for responsible AI deployment.
FLUX: Feedback Latency and Utilization Examination — Optimizing Real-Time AI Pipelines
The aim of this thesis is to extend the existing latency analysis of a psychomotor feedback engine within our existing MLOps pipeline [1] [2]. Building upon preliminary latency estimations, this thesis will focus on systematically evaluating each processing step in the pipeline, assessing both theoretical and practical contributions to the overall latency and throughput. By modeling and analyzing latency sources, the goal is to propose and validate optimization strategies that can improve real-time performance for sensor-based AI applications. A particular emphasis will be placed on the throughput of parallel data processing within the infrastructure to ensure timely and efficient feedback delivery.
Identification Techniques for Data-driven Feedback Loops in Manufacturing
This project aims to develop a methodology for the systematic selection and implementation of identifier systems in manufacturing, with a focus on ultrashort-pulsed (UKP) laser systems. By creating a robust identification framework tailored to manufacturing environments, the project will enhance data traceability, interoperability, and reusability within data-driven feedback loops, particularly in highly automated settings. The resulting methodology will support the seamless integration of measurement and control data, enabling advanced data tracking and analysis without compromising production information.
Mess to Mastery: User-centric redesign of psychomotor teaching experience
This master’s thesis builds on the Psychomotor Feedback Engine (PFE) and the IMPECT framework [1], [2], [3], aiming to improve the graphical user interface that teachers use to implement rules and feedback elements in psychomotor learning.
This thesis aims to address those usability issues by redesigning the teacher interface and integrating the IMPECT framework to introduce a more comprehensive feedback mechanism. The IMPECT framework introduces two core concepts: “feedback cards” and exercise stages. Feedback cards provide structured guidance on how exercises should be performed, while exercise stages break down training into a series of sequential tasks. Together, these elements can create a more dynamic and responsive feedback system, offering learners timely, actionable feedback during their training sessions.
Leveraging LLMs for Mathematical Optimization on the Example of Supply Chains
TLM: Bridge the Modality Gap between Transcriptome and Natural Language
This thesis focuses on developing a Transcriptome-Language Model (TLM) to effectively bridge the modality gap between transcriptomic data and natural language text. You will explore advanced models for transcriptomic data representation, and cross-modal learning techniques aligning transcriptomic and textual modalities. This model will be evaluated in tasks such as zero-shot cell property classification and text generation.
This thesis is co-supervised by Sikander Hayat and Rafael Kramann, Department of Medicine II, University Hospital Aachen.
Please send your application to Yongli Mou, M.Sc. (mou@dbis.rwth-aachen.de) and CC. Dr. Sikander Hayat (shayat@ukaachen.de)
Code-Based API Generation and Integration for Graph Analysis Algorithms
Knowledge Graph Construction for German Law Documents
This project aims to develop a comprehensive knowledge graph that represents German law documents, including cases and statutes. By creating an ontology tailored to the legal domain and leveraging automated annotation techniques, the project will transform unstructured legal text into structured data that can be queried. This knowledge graph will support legal research, enhance information retrieval, and enable semantic analysis of German legal documents.