Hello, people from the future! I am Sujan, a PhD student at RIT in Computing and Information Sciences. My PhD advisor is Dr. Ashique KhudaBukhsh. My research is dedicated to extracting social insights through NLP tools and developing reliable AI systems. My work has been recognized at top AI conferences, including IJCAI and EMNLP. Beyond academia, my role as an educational content creator at Normalized Nerd allows me to share my knowledge in machine learning and data science, impacting a global audience of over 100,000 subscribers.
Latest Updates
- My YouTube channel got featured on RIT News.
- Our work on Rater Cohesion was accepted at EMNLP 2024 (Findings).
- Our work on RLAIF for lightweight LLMs was featured on the Apple ML Research website.
- RIT News featured our work on bias auditing LLMs.
- I analyzed the stories from Panchatantra (an ancient Indian text) using AI and made infographics!
Selected Papers
- RLAIF for Code Generation with API-usage in Lightweight LLMs.
(ACL NLRSE 2024). - Down the Toxicity Rabbit Hole: A Novel Framework to Bias Audit Large Language Models.
(IJCAI 2024). - Vicarious Offense and Noise Audit of Offensive Speech Classifiers: Unifying Human and Machine Disagreement on What is Offensive.
(EMNLP 2023). - A Murder and Protests, the Capitol Riot, and the Chauvin Trial: Estimating Disparate News
Media Stance.
(IJCAI 2022).