I'm currently a Master's student in Computer Science at the University of Maryland, Baltimore County. I'm in the LARA Lab , advised by Dr. Lara Martin, working on affective and personalized text generation for AAC applications.
My research interests span interdisciplinary work in Human-AI Collaboration for Accessibility across multiple modalities – Vision, Text and Audio – as well as ensuring that the models used are also Explainable and Trustworthy.
See About for more details on my research interests and background.
News
I successfully defended my thesis on Connecting Language and Emotion in LLMs for Human-AI Collaboration.
The preprint of my paper on Emotion Representation Alignment between LLMs and Humans is out.
My paper on Gold Nanoparticle imaging and analysis published in ACS Omega (IF: 3.7)!
Three papers from my previous lab published in PLoS ONE (IF: 3.1), IJEST (IF: 3.0) and TRENG (IF: 5.4). I had a relatively small contribution to each.
Won Best First Time Hackers Category at HackUMBC 2024 with Naren . We built Valkyrie, a system that parses video game controls with a VLM, then uses a LLM to convert speech to commands.
My Paper on Gold Nanoparticle counting was published in RSC Analyst (IF: 4.2)!
Joined the LARA Lab , working on affective and personalized text generation with Dr. Lara Martin !
Started MS in Computer Science at the University of Maryland, Baltimore County!
One Paper on Audio Processing presented at MICAI 2021. Shout-out to my co-author Ishrat for leading!
Joined the Nanotechnology, IoT and Applied ML Group at Brac University working with Dr. Abu Mohsin .
One Paper on Fuzzy Logic published (first pub!) in Inderscience IJIIDS (IF: 1.4).
Finished my BS in Computer Science and Engineering at North South University!
Highlighted Papers

GPT’s Devastated 😭 and LLaMA’s Content 😌: Emotion Representation Alignment in LLMs for Keyword-based Generation
S. H. Choudhury, A. Kumar and L. J. Martin
arXiv

Quantifying Monomer-Dimer Distribution of Nanoparticles from Uncorrelated Optical Images Using Deep Learning
S. H. Choudhury and A. S. M. Mohsin
ACS Omega

Label-Free Quantification of Gold Nanoparticles at the Single-Cell level using MC-CNN
S. H. Choudhury and A. S. M. Mohsin
Analyst

A Comparative Study on Approaches to Acoustic Scene Classification using CNNs
I. J. Ananya, S. Suad, S. H. Choudhury and M. A. Khan
MICAI 2021

Predicting the Possibility of COVID-19 Infection Using Fuzzy Logic System
S. H. Choudhury, A. J. Aurin, T. A. Mitaly and R. M. Rahman
Int'l Journal of Intelligent Information & Database Systems