Research Interests
My research interests are deep learning and computer vision, focusing on vision alignment. I work on deep learning model architectures, training, optimisation, and understanding. I also enjoy working on projects in mechanistic interpretability and reinforcement learning, with an emphasis on vision models, but LLMs are very cool as well :)
Publications
Integrating self-configuring and foundational deep-learning segmentation models for identifying the anal sphincter complex and perianal fistulas on pelvic MRI
Sridharan, A.,
DeSilivio, Tom.,
Viswanath, S.
SPIE: Medical Imaging, 2025
Integrating multi-plane and multi-region radiomic features to predict pathologic response to neoadjuvant chemoradiation in rectal cancers via pre-treatment MRI
DeSilvio, T.,
Sridharan, A.,
Viswanath, S.
SPIE: Medical Imaging, 2023
Projects
I have contributed to multiple projects, primarily associated with INVent Labs. Additionally, my projects, which focus on Mechanistic Interpretability topics, Vision models implemented in C, and Flash Attention deployment, are available on GitHub. Please feel free to reach out to me if you want to collaborate on any project - I am interested in literally anything, so any project would be fun for me to work on.
