Curriculum Vitae
Education, research, teaching, service, and awards at a glance. Download the PDF version.
- General Information
- Research Interests
- Education
- Research Experience
- Publications
- Teaching Experience
- Honors and Awards
- Academic Service
- Talks and Presentations
- Competitive Programming
- Certifications
- Technical Skills
General Information
- Full Name
- Shahriyar Zaman Ridoy
- shahriyar.zaman01@gmail.com
- Website
- shahriyar-zaman.github.io
- Google Scholar
- scholar.google.com/citations?user=Tj7LsaQAAAAJ
- GitHub
- github.com/shahriyar-zaman
- linkedin.com/in/shahriyar-zaman
- Location
- Dhaka, Bangladesh
- Languages
- English, Bangla
Research Interests
Grounded Multimodal Reasoning
- Geo-temporal, geo-spatial, and spatial reasoning in vision-language models
- Diagnostic benchmarks that separate genuine reasoning from shortcut exploitation
Formal and Symbolic Reasoning
- LLMs as designers of executable computational systems: grammars, pushdown automata, Turing machines
- Behavioral correctness evaluation for generated formal artifacts
Trustworthy and Culturally Grounded Evaluation
- Hallucination detection, adversarial robustness, and efficiency of foundation models
- Culturally grounded evaluation for Bengali and other underrepresented languages
Education
B.Sc. in Computer Science and Engineering, Summa Cum Laude
North South University, Dhaka, Bangladesh
- CGPA 3.88 / 4.00 (top 2%); specialization GPA 4.00 / 4.00; 75% merit scholarship.
- Specialization: Natural Language Processing, Machine Learning.
- Thesis
- Context-Aware Data Cleaning Pipeline for Clinical Text and Low-Resource Bangla.
- Best Paper Award, IEEE International Conference on Intelligent Systems (IS) 2024.
- Journal extension published in SN Computer Science (Springer Nature), 2025.
Research Experience
Research Intern
Vanderbilt University (Remote)
Research Assistant
Cognitive Agents and Interaction Lab, University of Dhaka (Remote)
- Supervisor: Dr. Md Mosaddek Khan.
- Symbolic and formal reasoning in LLMs: executable grammars, pushdown automata, and Turing machines.
- Two papers accepted to Findings of EMNLP 2026 (one as first author); one paper under review at ACL Rolling Review.
Research Collaborator
Qatar Computing Research Institute (QCRI) (Remote)
- Supervisor: Dr. Md Rizwan Parvez.
- Evaluation protocols for geo-temporal and spatial reasoning in vision-language models.
- Co-first author of TimeSpot (ICML 2026); co-author of SpatiaLab (ICLR 2026); lead author of a geo-reasoning survey accepted to AACL-IJCNLP 2026.
Research Collaborator
Apurba-NSU R&D Lab, North South University
- Supervisors: Dr. Nabeel Mohammed and Dr. Shafin Rahman.
- Model compression (knowledge distillation) for LLMs and efficient vision-language reasoning.
- Joint first author of CAPSTONE, accepted to the EMNLP 2025 Industry Track.
Research Intern
Artificial Intelligence Institute, University of South Carolina (AIISC) (Remote)
- Collaborators: Prof. Amitava Das (BITS Pilani, formerly AIISC), Aman Chadha (AWS), Vinija Jain (Meta AI).
- Hallucination detection for text-to-image generation with Qwen-VL and LLaVA on MS-COCO and synthetic T2I benchmarks.
- Speculative-decoding defenses against adversarial attacks on diffusion models.
Research Assistant (NLP)
North South University
- Supervisor: Dr. Rashedur M. Rahman.
- Context-aware Bengali text-cleaning pipeline for transformer models: up to +4% accuracy on three benchmarks (SN Computer Science).
- Efficient clinical-text preprocessing pipeline: up to 17% faster training (Best Paper Award, IEEE IS 2024).
Publications
- Full list with abstracts and links on the Publications page.
- Main conferences: ICML 2026 (co-first author), ICLR 2026, Findings of EMNLP 2026 (first author and co-author), ACM FAccT 2026 (co-first author), AACL-IJCNLP 2026 (two main-conference papers and one Findings paper; co-first author on two), EMNLP 2025 Industry Track (co-first author), IEEE BigData 2024 (first author), IEEE IS 2024 (first author, Best Paper Award).
- Journal: SN Computer Science (Springer Nature), 2025.
- Workshops: ICML 2026 (AI for Math; Human-AI Co-Creativity), EurIPS 2025 (MMRL4H).
Teaching Experience
Lab Instructor
Department of Electrical and Computer Engineering, North South University
- CSE 225L Data Structures and Algorithms Lab; CSE 115L Programming Language I Lab.
- Lab Instructor Coordinator for CSE 225L: manual updates, problem-set design, standardized lab operations.
Teaching Assistant
Department of Electrical and Computer Engineering, North South University
- CSE 225 Data Structures and Algorithms; CSE 215 Programming Language II (OOP).
- Mentored 160+ students per semester; tutorials, grading, and detailed feedback.
Honors and Awards
- Best Paper Award, IEEE International Conference on Intelligent Systems (IS), Varna, Bulgaria.
- Summa Cum Laude, B.Sc. in Computer Science and Engineering, North South University (top 2%).
- Honorable Mention, ICPC Asia Dhaka Regional Online Preliminary.
- 75% Merit Scholarship for academic excellence, North South University.
Academic Service
- Reviewer, ICLR 2027.
- Reviewer, WACV 2027.
- Reviewer, NeurIPS 2026 Creative AI Track.
- Reviewer, NeurIPS 2026 AI4MetaScience Workshop.
- Reviewer, NeurIPS 2026 AIWILD Workshop.
- Reviewer, NeurIPS 2026 SLM-Agents Workshop.
- Reviewer, AI for Math Workshop at ICML 2026.
- Reviewer, ACL 2025 Student Research Workshop.
Talks and Presentations
ICML 2026, Seoul, South Korea
- Presented TimeSpot (main track) and workshop papers at AI for Math and Muslims in ML.
Competitive Programming
Programming Contestant, NSU Problem Solvers
- ICPC, Codeforces, AtCoder, Meta Hacker Cup, Vjudge.
- ICPC Asia Dhaka Online Preliminary: Honorable Mention.
- Solved roughly 1,000 problems across platforms; mentored peers in data structures and algorithms.
Certifications
Cohere Labs Open Science Community ML Summer School
Cohere Labs, with sessions from researchers at Meta, DeepMind, and INRIA
Supervised Machine Learning: Regression and Classification
Coursera (Andrew Ng's Machine Learning Specialization)
The Short Course on Data Science
Prof. Jennifer Widom (Stanford), Instructional Odyssey at NSU
Technical Skills
Languages
- Python
- C++
- C
- Java
- SQL
Machine Learning
- PyTorch
- TensorFlow / Keras
- Hugging Face Transformers
- scikit-learn
Data and Tooling
- NumPy / Pandas
- Matplotlib
- Git
- Docker
- LaTeX