"Let's advance the evaluation and explainability of language models to create AI tools that are both understandable and impactful across domains."
Ayoub Bagheri is an Associate Professor with promotion rights (ius promovendi), of Natural Language Processing (NLP) in the Department of Methodology and Statistics at Utrecht 木瓜福利影视. His research interests lie in fundamental and applied NLP, particularly in the evaluation and explainability of language models to develop transparent, reliable, and user-centred NLP systems.
Research interests:
- Explainable and Trustworthy AI
Developing methods to make NLP models transparent, interpretable, and accountable in high-stakes applications. - Language Model Evaluation
Designing benchmarks and methodologies to assess generalization, fairness, and stability of large language models. - Personalized NLP
Investigating how AI systems can align with individual user needs, cultural backgrounds, and cognitive styles. - Multimodal Understanding
Exploring how language models can reason over text, speech, and visual input across diverse languages and dialects. - Human-Centered NLP
Bridging computational modeling with human behavior and values, aiming for inclusive and socially responsible AI.
Ayoub leads the Natural Language and Text Processing () lab at M&S.
Links: