Md Toki Tahmid

Princeton University

Princeton, NJ 08544

sharifulislamtoki@gmail.com

I am a first-year PhD student in the Department of Computer Science at Princeton University, advised by Professor Mona Singh. My research lies at the intersection of machine learning and computational biology, with a focus on understanding protein function and regulation, RNA biology, and genomics through biological language models and graph neural networks.

I am broadly interested in building foundation models for multi-modal biomedical data — learning representations that capture the structure, function, and regulatory logic of biological sequences and networks. A key goal of my work is to make these models interpretable and clinically actionable.

Before joining Princeton, I completed my B.S. in Computer Science and Engineering from Bangladesh University of Engineering and Technology (BUET) with a CGPA of 3.98/4.00, working under Prof. Md. Shamsuzzoha Bayzid. I also had the privilege of interning at the Computational Connectomics Lab at MIT (with Nir Shavit and Lu Mi) and XuLab at Carnegie Mellon University (with Min Xu).

I am open to collaborations — feel free to reach out if our interests align!

selected publications

Nat. Commun. Biol.

TransBind allows precise detection of DNA-binding proteins and residues using language models and deep learning

Md. Toki Tahmid, A.K.M. Mehedi Hasan, and Md Shamsuzzoha Bayzid

Nature Communications Biology, Nov 2024

Accepted

Bib

@article{tahmid2024transbind,
  title = {TransBind allows precise detection of {DNA}-binding proteins and residues
                   using language models and deep learning},
  author = {Tahmid, Md. Toki and Hasan, A.K.M. Mehedi and Bayzid, Md Shamsuzzoha},
  journal = {Nature Communications Biology},
  year = {2024},
  month = nov,
  publisher = {Nature Publishing Group},
  note = {Accepted}
}

NeurIPS ENLSP

BiRNA-BERT: Adaptive Tokenization for Efficient RNA Language Modeling

Md Toki Tahmid, H.A.Z Sameen Sagir, Sazan Mahbub, and 2 more authors

In NeurIPS 2024 Workshop on Efficient Natural Language and Speech Processing, Sep 2024

\textbfSpotlight Paper · Under Major Revision at \textitNature Communications Biology · Collaboration: University of California, Riverside

Bib

@inproceedings{tahmid2024birnabertworkshop,
  title = {{BiRNA-BERT}: Adaptive Tokenization for Efficient {RNA} Language Modeling},
  author = {Tahmid, Md Toki and Sagir, H.A.Z Sameen and Mahbub, Sazan and Dong, Yue and Bayzid, Md Shamsuzzoha},
  booktitle = {NeurIPS 2024 Workshop on Efficient Natural Language and Speech Processing},
  year = {2024},
  month = sep,
  note = {\textbf{Spotlight Paper} · Under Major Revision at \textit{Nature Communications Biology} · Collaboration: University of California, Riverside},
}

NeurIPS

Structure Matters: Deciphering Neural Network’s Properties from its Structure

Shashata Sawmya, Md Toki Tahmid, Gourab Saha, and 3 more authors

In NeurIPS 2024 Workshop: Symmetry and Geometry in Neural Representations, Nov 2024

Joint Co-First Author · Collaboration: Computational Connectomics Lab, MIT

Bib

@inproceedings{sawmya2024structurematters,
  title = {Structure Matters: Deciphering Neural Network's Properties from its Structure},
  author = {Sawmya, Shashata and Tahmid, Md Toki and Saha, Gourab and Saha, Arpita and Shavit, Nir N and Mi, Lu},
  booktitle = {NeurIPS 2024 Workshop: Symmetry and Geometry in Neural Representations},
  year = {2024},
  month = nov,
  note = {Joint Co-First Author · Collaboration: Computational Connectomics Lab, MIT},
}