Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
Slot Machines to Agent Armies
Published:
This one is primarily me regurgitating theory so that I don’t keep asking questions like ‘wait, what was TD learning again?’
projects
Autonomous Underwater Vehicle (AUV) - IITB
A student team which develops underwater vehicles capable of performing naval tasks, competing annually at RoboSub, San Diego
Bachelor’s Thesis-I: Hallucination detection of LLMs
Collaborative research with Adobe India submitted to EMNLP 2024.
Bachelor’s Thesis-II: Unsupervised Segmentation of Agricultural Crop Fields
Automated Normalized difference vegetation index (NDVI) calculation for TIH-IOT, IITB under the supervision of Prof. Rajbabu Velmurugan.
publications
An algorithmic recipe to construct Lyapunov functions for continuous vector fields
Published in Gold Coast, Australia, 2024
In this work, we introduce a novel and computationally feasible approach to construct Lyapunov functions for equilibria of nonlinear continuous vector fields.
Recommended citation: S. Chattopadhyay, R. Gupta, P. Paruchuri and D. Chatterjee,“An algorithmic recipe to construct Lyapunov functions for continuous vector fields,” 2024 Australian & New Zealand Control Conference
Download Paper | Download Slides
Human-centered artificial intelligence framework for overdose mortality surveillance
Published in American Journal of Public Health, 2025
We propose a human-centered AI framework for overdose mortality surveillance using evidence extraction from Medicolegal Death Investigation reports.
Recommended citation: K. Hochstatter, S. Muresan, R. Gupta, T. Nadel, A. Jangra, Y. Li, C. Yang, N. D'Anna, H. Johnson, J. Gryczynski, N. El-Bassel, J. Graham, "Human-centered artificial intelligence framework for overdose mortality surveillance," American Journal of Public Health.
Consistency Is the Key: Detecting Hallucinations in LLM Generated Text By Checking Inconsistencies About Key Facts
Published in Findings of IJCNLP-AACL 2025, Mumbai, India, 2025
We present CONFACTCHECK, a hallucination detection method that checks for consistency in LLM responses to factual probes, requiring no external knowledge base.
Recommended citation: R. Gupta, P. H. Panicker, S. Bhatia, G. Ramakrishnan, "Consistency Is the Key: Detecting Hallucinations in LLM Generated Text By Checking Inconsistencies About Key Facts," Findings of IJCNLP-AACL 2025.
Download Paper
teaching
Differential Equations & Calculus
Undergraduate course, IIT Bombay, Mathematics Department, 2023
I have had the pleasure to be a Teaching Assistant (TA) for 120+ freshmen and sophomores at IIT Bombay on four distinct occasions. My experience about one of them can be found in this blog curated for the Electrical Engineering Department Academic Mentorship Program, IITB. Courses I have assisted:
- MA108: Differential Equations - I
- MA207: Differential Equations - II
- MA111: Calculus (twice)
Online Learning and Optimization
Graduate Course, IIT Bombay, Electrical Engineering (EE) Department, 2024
In my last semester at IIT Bombay, I TA’ed a graduate course comprising of 60+ students under the supervision of Prof. Jayakrishnan Nair, EE, IITB. The duties involved grading the assignments, mid-term and end-term examinations!
Programming and Problem Solving
Graduate Course, Columbia, Computer Science (CS) Department, 2024
In Fall 2024, I served as the teaching assistant for the course W4444 Programming and Problem Solving, where I developed simulators (1, 2) and assisted Prof. Kenneth Ross in running the course.