EvalMORAAL: Interpretable Chain-of-Thought and LLM-as-Judge Evaluation for Moral Alignment in Large Language Models
IOPS Winter Conference 2025, Utrecht University — December 2025. Best Oral Presentation Award.
Research
Research
I work on explainable NLP, cultural fairness in large language models, and human-AI collaboration. PhD from Utrecht University.
01Themes
What I work on
Explainable NLP & interpretability
I build NLP models whose decisions domain experts can inspect and question. This covers post-hoc interpretability methods, token-level explanations, and transparent pipelines for sensitive classification tasks in the social sciences. All of this work follows open-science practice: code, data, and paper materials are released so results can be reproduced and extended.
Cultural fairness & moral reasoning in LLMs
Large language models encode moral judgments, but not everyone's. I study how LLMs represent cross-cultural variation in moral judgments and societal norms, and how to evaluate their alignment across cultures — including EvalMORAAL, an interpretable chain-of-thought and LLM-as-judge framework for moral alignment.
Human-AI collaboration & annotation reliability
When can we trust an LLM's annotations and explanations, and when do we still need a human? I compare model outputs against human rationalizations, measure demographic bias in annotation, and study how explanations should be written to match human expectations. A related line uses reinforcement learning to extract decision rules from human evaluations.
02Grants & awards
Funding and recognition
- Dec 2025 ADS grant, €5,000 — "Optimizing the CV Priority Sorter for Fair and Efficient Prioritization of Academic CVs", with Dr. Robert A. Bagheri, Dr. Georg Krempl, and Jeroen Sparla.
- Dec 2025 Best Oral Presentation Award — IOPS Winter Conference 2025, for work on cultural moral alignment in LLMs. Event details.
- Jul 2025 ADS grant, €5,000 — "Rule Extraction from Human Evaluation Using LLMs and Reinforcement Learning", with Dr. Anastasia Giachanou and Dr. Shihan Wang.
- Mar 2025 CUCo Spark grant, €9,000 — "ReDOSE: Towards a More Circular Pharmaceutical Industry using AI", with Dr. Negin Salimi (WUR), Dr. Shayegheh Ashourizadeh (WUR), Dr. Duru Bayram (TU/e), and Dr. Saeed Arbabi (UMC Utrecht).
- 2025 LERU Doctoral Summer School — selected as representative of the Faculty of Social and Behavioural Sciences, Utrecht University, for the summer school on AI in Copenhagen.
- Dec 2024 ENFIELD grant, €14,400 — European Lighthouse to Manifest Trustworthy and Green AI, for "User Perspectives on Explainable AI".
- Dec 2023 €7,500 Applied Data Science grant for “Assessing Reliability of Annotations in the Context of Model Predictions and Explanations”, with Dr. Pablo Mosteiro Romero, Dr. Anastasia Giachanou, and Prof. Dr. Massimo Poesio.
- Aug 2022 Scholarship for admission to the Explainable AI (XAI) Summer School, TU Delft.
- 2019 Best Paper Award, Second National Conference of the Iranian System Dynamics Society.
03Talks & posters
Presentations
Assessing the Reliability of Annotations in the Context of LLM Predictions and Explanations
6th NLTP Content Meeting Session, Utrecht — February 2025.
Novel Approaches in Financial Fraud Detection: Hybrid Machine Learning and Uncertainty-Based Deep Learning
BNAIC/BeNeLearn 2024, Utrecht — November 2024. Presented as Communications Chair of the conference.
Towards Explainable AI-Generated Text Detection Using Ensemble and Combined Model Training
39th IOPS Winter Conference, Amsterdam — December 2023.
A Journey on Explainable Natural Language Processing (NLP) in Social Science Applications
AI-lab Users Meeting (ASReview), Utrecht University — November 2023.
AI-Generated Text Detection Using Ensemble and Combined Model Training
33rd Meeting of Computational Linguistics in the Netherlands (CLIN33), Antwerp — October 2023.
A Journey on Explainable Natural Language Processing (NLP) in Social Science Applications
Human Data Science Content-Oriented Meeting, Utrecht University — October 2023.
Towards Explainable Sexism Detection in Social Media: An Ensemble Approach with Human Rationalizations
33rd Meeting of Computational Linguistics in the Netherlands (CLIN33), Antwerp — September 2023.
Introduction to Transformers, BERT and Explainable NLP
Skills Lab, M&S Research Hay Day 2023, Utrecht University — June 2023. With Dr. Huyen Nguyen and Daniel Anadria.
04Collaboration
Work with me
I am open to research collaborations in explainable NLP, cultural fairness in LLMs, annotation reliability, and applied machine learning for the social sciences. If our interests overlap, email me at hadi.mohammadi@outlook.com.