Sujoy Nath

About Me

I am currently working at the IITD, on an Agentic AI project funded by Microsoft Research India (MSRI), co-supervised by Prof. Tanmoy Chakraborty and Principal Engineer at MSRI Dr. Akshay Nambi. I am primarily focusing on developing more trustworthy AI agents by enhancing their safety mechanisms and reasoning capabilities.

Previously, I also have had the opportunity to work extensively with Prof. Swagatam Das at the Indian Statistical Institute (ISI), where we developed methods not only for detecting hallucinations but also for mitigating hallucination and safety alignment in LLMs resulting in publications at top-tier venues.

My research interests span across trustworthy AI systems, LLM interpretability and safety, agentic reasoning, and AI applications in healthcare. I am particularly drawn to developing AI systems that are not only powerful but also reliable, transparent, and beneficial for society.

Research Interests

Trustworthy AI

Developing reliable, interpretable, and safe AI systems with focus on transparency and accountability.

LLM Interpretability & Safety

Investigating methods to understand and improve the safety mechanisms of large language models.

Agentic Reasoning

Exploring autonomous decision-making capabilities in AI systems and multi-agent interactions.

AI for Medical Domain

Applying AI technologies to healthcare challenges with emphasis on reliability and explainability.

Natural Language Processing

Advancing NLP techniques for better human-AI interaction and understanding.

Multimodal AI Systems

Developing AI systems that can process and understand multiple types of data simultaneously.

Publications

From Complexity to Clarity: Transforming Chest X-ray Reports with Chained Prompting

Sujoy Nath, Arkaprabha Basu, Kushal Bose, Swagatam Das

AAAI 2025 Student Abstract (Ranking: A*)

DOI

HalluShift: Measuring Distribution Shifts towards Hallucination Detection in LLMs

Sharanya Dasgupta, Sujoy Nath, Arkaprabha Basu, Pourya Shamsolmoali, Swagatam Das

IJCNN 2025 (Ranking: B)

arXiv

Hallucination Mitigation & Safety Alignment in LLMs (Title Representational)

Under Review

Sharanya Dasgupta, Arkaprabha Basu, Sujoy Nath, Swagatam Das

CIKM 2025 (Ranking: A)

Hierarchical Hallucinations detection in MLLMs (Title Representational)

Under Review

Sujoy Nath, Arkaprabha Basu, Sharanya Dasgupta, Swagatam Das

ICVGIP 2025

Experience

Research Experience

Research Assistant

Present

Indian Institute of Technology (IIT), Delhi

Supervisor: Prof. Tanmoy Chakraborty

Working on an Agentic AI project funded by Microsoft Research India (MSRI), co-supervised by professor Dr. Tanmoy Chakraborty and principal engineer at MSRI Dr. Akshay Nambi.
I am primarily focusing on making AI agents more trustworthy by evaluating their safety and improving reasoning capabilities.

Research Collaborator

June 2024 - June 2025

Indian Statistical Institute (ISI), ECSU

Supervisor: Dr. Swagatam Das

HalluShift: Developed novel approach for detecting factual hallucinations in LLM outputs by analyzing distributional shifts in internal state space and token probabilities. Accepted at IJCNN 2025
Medical Report Generation: Created simplified MRG system using Gemini-1.5-Flash, fine-tuned LLaMa models, and introduced CPMK-E scoring method. Accepted as Student Abstract at AAAI 2025
Image Captioning Pipeline: Built system using BLIP embeddings and CerberusDet (YOLOv8) for object detection, fine-tuned multiple LLMs (LLaMa 3.1/2, Mistral 7B, Phi-2) with comprehensive evaluation metrics

Summer Intern

March 2024 - April 2024

Defence Research and Development Organisation (DRDO), DEBEL Lab

Developed comprehensive dataset using IMU sensor data from Xsens, capturing walking patterns of 20 subjects for motion analysis
Researched human gait pattern analysis using ensemble learning methods for real-time prediction systems in lower limb prosthetics

Industry Experience

Machine Learning Engineer Intern

October 2023 - May 2025

Geogo Techsolutions

Face Recognition System: Built end-to-end face comparison and similarity search solution achieving 98% accuracy vs. Amazon Rekognition for employee authorization and verification
Speech Recognition: Engineered ASR model using Transformer architecture for Hindi-English mixed data, achieving 34% WER through innovative audio preprocessing (STFT, Conv1D), deployed for insurance call transcription

Education

Bachelor of Technology

Computer Science and Business System

Netaji Subhash Engineering College

August 2021 - June 2025

CGPA: 8.66/10 (3.46/4)

Achievements

Parayas 2k24

2nd Position - Software Segment

Team Lead | May 2024

Kavach 2023

Top 5 Finalist

Team Lead | Ministry of Education