Currently working on:
- MLOps
- Visual large language models
- Audio-in large language models for rare languages
- Morphological analysis using large language models
- Sequence-to-sequence transformers
- Investigate, design, use, and run deep learning approaches needed for the research projects.
- Applying state-of-the-art deep learning methods in semantic representation.
Background
- Ph.D. in Computer Science, Boğaziçi University, Computer Engineering (2020)
- MSc. in Software Engineering, Boğaziçi University (2007)
- BSc. in Computer Engineering, Karadeniz Technical University (2005)
Publications
- Ahmet Yıldırım, Kristin Hagen, Dag Trygve Truslew Haug, Automatic Lemmatisation for Norwegian, Proceedings of the Workshop on Structured Linguistic Data and Evaluation (SLiDE 2026), pages 93–103, May 11, 2026, ELRA Language Resources Association (ELRA).
- Ahmet Yıldırım and Dag Trygve Truslew Haug, Experiments in training transformer sequence-to-sequence DRS parsers, The 15th International Conference on Computational Semantics, IWCS 2023, Nancy, France
- Dag T. T. Haug, Jamie Y. Findlay, and Ahmet Yildirim, The long and the short of it: DRASTIC, a semantically annotated dataset containing sentences of more natural length, The 4th International Workshop on Designing Meaning Representation, DMR2023, Nancy, France
- Dag Haug, Ahmet Yildirim, Kristin Hagen, Anders Nøklestad, Rules and neural nets for morphological tagging of Norwegian - Results and challenges. 24th Nordic Conference on Computational Linguistics (NoDaLiDa), 425–435, Tórshavn, Faroe Islands, 2023
- Jamie Y. Findlay, Saeedeh Salimifar, Ahmet Yıldırım, Dag T. T. Haug, Rule-based semantic interpretation for Universal Dependencies. Sixth Workshop on Universal Dependencies (UDW, GURT/SyntaxFest 2023), 47–57, Washington, D.C., ACL, 2023