Arshak Rezvani

Hi, I’m Arshak Rezvani. I am currently a Master’s student in Information and Communication Engineering at Technical University of Darmstadt, with a background in Electrical Engineering (Bioelectric track) from Sharif University of Technology. I’m passionate about training neural networks to sample from complex data distributions :).

My recent work spans a variety of exciting research areas, including:

Generative AI & Diffusion Models
AI in Healthcare and Biomedical Applications
Interpertability of Large Language Models
Contrastive Learning & CLIP-style Multimodal Pretraining

selected publications

ICML

Interpretable EEG-to-Image Generation with Semantic Prompts

Arshak Rezvani^*, Ali Akbari^*, Kosar Sanjar Arani, and 3 more authors

2025

arXiv Bib Video Poster Website

@misc{rezvani2025interpretableeegtoimagegenerationsemantic,
  title = {Interpretable EEG-to-Image Generation with Semantic Prompts},
  author = {Rezvani, Arshak and Akbari, Ali and Arani, Kosar Sanjar and Mirian, Maryam and Arasteh, Emad and McKeown, Martin J.},
  year = {2025},
  journal = {Actionable Interpretability Workshop at International Conference on Machine Learning},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
  url = {https://arxiv.org/abs/2507.07157},
  equal_contribution = {Arshak Rezvani and Ali Akbari},
}

IEEE

DiffuseGaitNet: Improving Parkinson’s Disease Gait Severity Assessment With a Diffusion Model Framework

Arshak Rezvani, Nasrin Ravansalar, Mohammad Ali Akhaee, and 5 more authors

IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2025

DOI Bib

@article{11080070,
  author = {Rezvani, Arshak and Ravansalar, Nasrin and Akhaee, Mohammad Ali and Greenshaw, Andrew J. and Greiner, Russell and Mirian, Maryam S. and Yousefnezhad, Muhammad and McKeown, Martin J.},
  journal = {IEEE Transactions on Neural Systems and Rehabilitation Engineering},
  title = {DiffuseGaitNet: Improving Parkinson’s Disease Gait Severity Assessment With a Diffusion Model Framework},
  year = {2025},
  volume = {33},
  number = {},
  pages = {2858-2869},
  keywords = {Diffusion models;Diseases;Data models;Feature extraction;Transformers;Three-dimensional displays;Noise;Uncertainty;Training;Predictive models;Diffusion models;generative AI;transformers;attention-based networks;Parkinson’s disease;gait impairments;human action recognition;MDS-UPDRS},
  doi = {10.1109/TNSRE.2025.3589074},
}