Leonardo Ranaldi is an NLP Researcher focused on the development, analysis, and interpretability of neural language models. His work sits at the intersection of Computer Science and Linguistics, contributing to the evolution of machine learning techniques through a deep understanding of how artificial neural networks process language.
Currently, he is Research Associate at the University of Edinburgh.
Research Focus
His research focuses on the “how” and “why” of modern AI, specifically addressing:
- Interpretability & Explainability: Analysing linguistic phenomena learned by neural networks to move beyond “black-box” models.
- Model Refinement: Enhancing the multilingual capabilities and robustness of contemporary language models.
- Reasoning & Task Optimisation: Developing methodologies for Question Answering (QA), Neural Machine Translation (NMT), and complex reasoning tasks.
Academic Contributions
Leonardo has authored over 30+ publications in the field of Natural Language Processing. His research trajectory spans from foundational linguistic analysis to the practical enhancement of model performance in multilingual settings.
Professional Service
- Organiser
- MathNLP
- SemEval 2026 Task 11
- Senior Area Chair:
- EMNLP 25
- EACL 26
- Area Chair:
- ACL,
- EMNLP
- EACL
- NAACL
- Reviewers:
- NLP: AACL, NAACL, EACL, EMNLP, ACL, COLM
- AI/ML: UAI, AISTATS, NEURIPS, ICLR, ICML, AAAI
- Journal:
- NeuroComputing
- Transactions on Machine Learning Research
- Transactions on Big Data
- Transactions on Artificial Intelligence