Mehdi Yazdani-Jahromi

AI research scientist and PhD in Computer Science specializing in representation learning, with primary applications in computational biology and drug discovery. Industry experience at Johnson and Johnson and Microsoft, where I developed language models for mRNA and machine learning systems for large scale biomedical data.

Strong programming background in Python, and JavaScript, with deep experience in PyTorch, Transformers and graph neural networks. Published in NeurIPS, ICLR and Briefings in Bioinformatics, with work spanning mRNA modeling, drug target interaction prediction and algorithmic fairness.

Passionate about building transparent, biologically grounded AI systems and expanding the capabilities of machine learning to drive discovery in the life sciences.

News

Oct 22, 2025	I have successfully defended my PhD in Computer Science at UCF.
Sep 25, 2025	Our paper, “Equi-mRNA: Protein Translation Equivariant Encoding for mRNA Language Models,” has been accepted to NeurIPS 2025. We introduce Equi-mRNA, the first codon-level equivariant mRNA language model that explicitly encodes synonymous codon symmetries as cyclic subgroups of 2D Special Orthogonal matrix (SO(2)).
Jun 30, 2025	I have joined Microsoft Research as a Research Intern in the Health Futures group, where I’m working on AI-driven solutions for biomedical and clinical challenges.
Jan 22, 2025	Our paper, “HELM: Hierarchical Encoding for mRNA Language Modeling,” has been accepted to ICLR 2025. Our work introduces a novel approach to understanding mRNA sequences by leveraging hierarchical language modeling, outperforming existing baselines by ~8% across six downstream property prediction tasks. Check out the paper at: arXiv
Sep 25, 2024	Our paper, “Fair Bilevel Neural Network (FairBiNN): On Balancing Fairness and Accuracy via Stackelberg Equilibrium,” has been accepted to NeurIPS 2024. We introduce a novel bilevel optimization approach to enhance fairness in machine learning, achieving superior results on key datasets.

Latest Posts

May 18, 2025	Why Regularization Isn’t Enough: A Better Way to Train Neural Networks with Two Objectives

Selected Publications

Equi-mRNA: Protein Translation Equivariant Encoding for mRNA Language Models

Mehdi Yazdani-Jahromi , Ali Khodabandeh Yalabadi , and Ozlem Ozmen Garibay

Neurips 2025, 2025

Bib

@article{yazdani2025equi,
  title = {Equi-mRNA: Protein Translation Equivariant Encoding for mRNA Language Models},
  author = {Yazdani-Jahromi, Mehdi and Yalabadi, Ali Khodabandeh and Garibay, Ozlem Ozmen},
  journal = {Neurips 2025},
  url = {https://arxiv.org/abs/2508.15103},
  year = {2025}
}

Fair Bilevel Neural Network (FairBiNN): On Balancing fairness and accuracy via Stackelberg Equilibrium

Mehdi Yazdani-Jahromi , Ali Khodabandeh Yalabadi , AmirArsalan Rajabi , and 3 more authors

Neurips 2024, 2024

Bib

@article{yazdani2024fair,
  title = {Fair Bilevel Neural Network (FairBiNN): On Balancing fairness and accuracy via Stackelberg Equilibrium},
  author = {Yazdani-Jahromi, Mehdi and Yalabadi, Ali Khodabandeh and Rajabi, AmirArsalan and Tayebi, Aida and Garibay, Ivan and Garibay, Ozlem Ozmen},
  journal = {Neurips 2024},
  url = {https://arxiv.org/abs/2410.16432},
  year = {2024}
}

HELM: Hierarchical Encoding for mRNA Language Modeling

Mehdi Yazdani-Jahromi , Mangal Prakash , Tommaso Mansi , and 2 more authors

ICLR 2025, 2024

Bib

@article{yazdani2024helm,
  title = {HELM: Hierarchical Encoding for mRNA Language Modeling},
  author = {Yazdani-Jahromi, Mehdi and Prakash, Mangal and Mansi, Tommaso and Moskalev, Artem and Liao, Rui},
  journal = {ICLR 2025},
  url = {https://arxiv.org/abs/2410.12459},
  year = {2024}
}

FragXsiteDTI: Revealing Responsible Segments in Drug-Target Interaction with Transformer-Driven Interpretation

Ali Khodabandeh Yalabadi , Mehdi Yazdani-Jahromi , Niloofar Yousefi , and 3 more authors

In International Conference on Research in Computational Molecular Biology , 2024

Bib

@inproceedings{khodabandeh2024fragxsitedti,
  title = {FragXsiteDTI: Revealing Responsible Segments in Drug-Target Interaction with Transformer-Driven Interpretation},
  author = {Khodabandeh Yalabadi, Ali and Yazdani-Jahromi, Mehdi and Yousefi, Niloofar and Tayebi, Aida and Abdidizaji, Sina and Garibay, Ozlem Ozmen},
  booktitle = {International Conference on Research in Computational Molecular Biology},
  url = {https://arxiv.org/abs/2311.02326},
  pages = {68--85},
  year = {2024},
  organization = {Springer}
}

AttentionSiteDTI: an interpretable graph-based model for drug-target interaction prediction using NLP sentence-level relation classification

Mehdi Yazdani-Jahromi , Niloofar Yousefi , Aida Tayebi , and 4 more authors

Briefings in Bioinformatics, 2022