28/04/2026
We are delighted to share a recent achievement by our students, Shefayat E Shams Adib, Ahmed Alfey Sani, and Md Hasibur Rahman Alif, from the Department of Computer Science and Engineering, Islamic University of Technology (IUT).
They secured 3rd place on the All-Class Leaderboard and 4th place on the Positive-Class Leaderboard at the PsyDefDetect 2026 Shared Task, held at the ACL BioNLP Workshop 2026, one of the most prestigious venues in computational linguistics and clinical NLP research worldwide.
The Association for Computational Linguistics (ACL) is the premier international scientific and professional organization in the field of NLP and computational linguistics, and the BioNLP Workshop is one of its longest-running and most respected venues, bringing together top researchers from across the globe working at the intersection of language technology and biomedical and clinical sciences.
Out of 168 participants and 563 total submissions, Team LinguIUTics rose to the very top, showcasing the depth of competition and the significance of this achievement on a truly global stage. Their work on classifying psychological defense mechanisms in conversational text achieved an impressive F1-score of 0.4427 on the all-classes leaderboard and F1-score of 0.3917 on the positive-classes leaderboard, surpassing the previous state-of-the-art by 24.6%, a remarkable breakthrough in a highly challenging clinical NLP problem.
A special appreciation goes to their supervisor, Ajwad Abrar, Junior Lecturer, Department of CSE, IUT, whose mentorship and guidance were crucial in this success.
This achievement is a testament to the research excellence and innovation thriving at IUT CSE. We look forward to many more such milestones!
