thesis proposals
Bachelor's and Master's thesis topics I am available to supervise.
I am open to supervising Bachelor’s and Master’s theses on topics related to Natural Language Processing, Semantic Web, and fair/inclusive language technologies. Below are some directions I am currently interested in; feel free to get in touch to discuss a proposal of your own that fits these themes.
If you are interested in any of the topics below, or would like to propose your own idea, please contact me with a short description of your background and interests.
### Bias and uncertainty in Large Language Models Investigate how model uncertainty (e.g., token-level entropy, calibration) can be used as a proxy to detect social bias in the outputs of Large Language Models, across different pretraining corpora and model sizes. ### Neuro-symbolic integration of structured knowledge in NLP Explore methods for integrating structured knowledge (knowledge graphs, ontologies) with neural NLP pipelines, with the goal of improving factual grounding and reducing hallucinations. ### Representational and allocative bias in digital archives Extend prior work on bias detection in digital archives to new domains or languages, studying how historical and institutional biases are encoded and propagated through NLP systems trained on archival text. ### Perspectivist approaches to Hate Speech detection Design and evaluate models that account for annotator disagreement and diverse social perspectives when detecting Hate Speech, moving beyond majority-vote gold labels.