PhD Student Mohammed Bouri’s Paper Accepted at ACL 2025

Mohammed Bouri, PhD student at the College of Computing at UM6P, has had his research paper accepted at the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), a top-tier international conference in the field of Natural Language Processing (NLP) and ranked A* in the CORE classification.

Under the supervision of Prof. Adnane Saoud, his paper is titled “Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Matrix Bound Growth Approach.” This work introduces an original regularization framework that uses growth-bound matrices to control the influence of adversarial input perturbations. The method offers a principled way to improve both robustness and generalization in NLP systems—especially those vulnerable to word-level attacks.With this contribution, UM6P affirms its active role in cutting-edge NLP research, particularly in building resilient AI systems for multilingual and real-world applications.