Ibrahim Riza Hallac

Norwegian version of this page Department of Computer Science and Communication
Image of Ibrahim Riza Hallac
Norwegian version of this page Position
Postdoctoral Fellow
Contact Study place
Halden
Office nr.
D1-018a

 

I am a Postdoctoral Fellow in the Department of Computer Science and Communication at Østfold University College, Norway. My research focuses on natural language processing and machine learning, with particular emphasis on representation learning and the adaptation and evaluation of large language models (LLMs). Within the AI4AfL (Artificial Intelligence for Assessment for Learning) project, I develop and evaluate LLM-based approaches for automated essay scoring (AES) and feedback generation, including staged model adaptation and rubric-aligned evaluation. I received my PhD in Computer Engineering from Firat University, Turkey. During my graduate studies, I conducted research visits at Indiana University (USA) and LMU Munich (Germany). I have previously worked on neural network optimization for real-world applications and have taught courses on Machine Learning and the Ethics of AI and Algorithms.

 

Publications

  • Özgür, Faruk; Tuncer, Taner; Hallac, Ibrahim Riza & Kaya, Mehmet (2026). Analysis of the Recall Crisis and Semantic Robustness of Llm Detectors Under Hierarchical Deep Adversarial Rewriting Attacks. Proceedings - International Conference on Advanced Computer Information Technologies, ACIT. p. 1307–1313. doi: 10.1109/ACIT69831.2026.11700022. Full text in Research Archive
  • Qassi, Abdelaziz; Hallac, Ibrahim Riza & Ogul, Hasan (2026). Investigating the Relevance of NLP Benchmarks for Automated Essay Grading: A Multi-trait Evaluation of Pre-trained Transformers. Lecture Notes in Networks and Systems. ISSN 2367-3370. 1754, p. 137–151. doi: 10.1007/978-3-032-13153-9_11. Full text in Research Archive
  • Leseanu, Anda; Hallac, Ibrahim Riza; Ogul, Burcin Buket & Ogul, Hasan (2025). Revisiting Automatic Essay Assessment: A Relative Approach. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. p. 41–52. doi: 10.1007/978-3-032-08462-0_4. Full text in Research Archive

View all works in NVA

View all works in NVA

Published Aug. 1, 2024 10:52 AM - Last modified Feb. 9, 2026 11:43 AM