Tanvir Ahmed Sijan

Tanvir Ahmed Sijan

Prospective PhD Applicant

Hi! 👋🏻 I’m Tanvir Ahmed Sijan, currently pursuing an M.Sc. in Computer Science at Jahangirnagar University through a project-based program under the supervision of Dr. Md. Musfique Anwar at the Computational Intelligence, Data and Language Research Lab, with my research focused on Natural Language Processing. I completed my Bachelor’s degree in Civil Engineering from Bangladesh University of Engineering and Technology (BUET).

My research interests lie at the intersection of NLP and AI safety, focusing on human-AI alignment through human-centered evaluation, alignment methods, and LLM interpretability to develop safer and more trustworthy language models.

My updated CV is available here.

News

Latest Checkpoints

  • 📆 [August 2026 - September 2026]

    • 🎉 Our mechanistic interpretabilty paper on cross-format arithmetic neurons has been accepted at BlackboxNLP @ EMNLP 2026 which received unprecedented amount of submissions this year!

    • 🎉 Our paper on event detection from noisy text has been accepted at W-NUT @ EMNLP 2026

  • 📆 [July 2026]

    • 📢 Our mechanistic interpretabilty paper on cross-format arithmetic neurons is now available on arXiv; currently under review
    • 🏆 BanglaSocialBench has received Best Paper Award at ACL SRW 2026!
    • 📢 Our paper on evaluating event detection in noisy text is now available on arXiv; currently under review

  • 📆 [March - April 2026]

    • 🎉 BanglaSocialBench has been accepted for Oral Presentation at ACL 2026 SRW!

    • 📢 BanglaSocialBench preprint available on arXiv; currently under review.

  • 📆 [May 2025]

  • 📆 [March 2025]

Peer-Reviewed Publications

Selected Publications and Preprints

* indicates equal contribution.

BlackboxNLP @ EMNLP 2026
Are Arithmetic Heuristic Neurons Form-Invariant? A Mechanistic Analysis of Symbols, Text, and Code in LLMs thumbnail
Are Arithmetic Heuristic Neurons Form-Invariant? A Mechanistic Analysis of Symbols, Text, and Code in LLMs
Sharath Naganna*, Tanvir Ahmed Sijan*, Uddipta Kalita
In Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, co-located with EMNLP 2026
Abs PDF Code
W-NUT @ EMNLP 2026
Beyond Clean Text: Evaluating Encoder and Decoder Robustness for Bangla Event Detection in Noisy Text thumbnail
Beyond Clean Text: Evaluating Encoder and Decoder Robustness for Bangla Event Detection in Noisy Text
Tanvir Ahmed Sijan, S. M Golam Rifat, Nayeemul Islam, Md. Musfique Anwar
In Proceedings of the 11th Workshop on Natural User-generated Text, co-located with EMNLP 2026
Abs PDF Code
SRW @ ACL 2026
BanglaSocialBench: A Benchmark for Evaluating Sociopragmatic and Cultural Alignment of LLMs in Bangladeshi Social Interaction thumbnail
BanglaSocialBench: A Benchmark for Evaluating Sociopragmatic and Cultural Alignment of LLMs in Bangladeshi Social Interaction
Tanvir Ahmed Sijan, S. M Golam Rifat, Pankaj Chowdhury Partha, Md. Tanjeed Islam, Md. Musfique Anwar
In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
Abs PDF Video Code
🏆 Best Paper Award
TRB 2027
Assessing the Visual and Built Environment Risk on Pedestrian-Vehicle Crash Severity in Absence of Detailed Roadway Inventory: A Deep Learning Based Framework Harnessing Public Geospatial Data thumbnail
Assessing the Visual and Built Environment Risk on Pedestrian-Vehicle Crash Severity in Absence of Detailed Roadway Inventory: A Deep Learning Based Framework Harnessing Public Geospatial Data
Tanvir Ahmed Sijan, Md Asif Raihan, Sk. Md. Mashrur, Md. Shamsul Hoque
Accepted at the 106th Transportation Research Board (TRB) Annual Meeting, Washington, D.C., 2027
Abs PDF

Honors & Awards

Highlights