What is your role at DECaDE?
At DECaDE, I serve as a Research Fellow, leveraging Multimodal Large Language Models to tackle the critical issue of misinformation on social media. My work centers on verifying digital content by integrating insights from diverse news sources, ensuring that misinformation is identified and addressed with precision.
Tell me about your area of research?
My research is at the intersection of Computer Vision and Natural Language Processing, where I focus on mitigating bias in pretrained models through the use of external knowledge. By applying these techniques to multimodal content verification, I aim to create more reliable tools for authenticating online information in real time, empowering users to differentiate between fact and falsehood.
What attracted you to working at DECaDE?
DECaDE presents an exceptional platform for developing innovative solutions to one of the most pressing challenges of our time: misinformation. With my background in verifying digital content and a strong commitment to real-world solutions, DECaDE provides the resources and collaborative environment necessary to elevate my research. Ultimately, the fight against misinformation is not just a professional endeavour for me—it’s a personal mission.
What is your background?
I hold a bachelor’s and master’s degree in Computer Science from Calcutta University, followed by a master’s and PhD from the Indian Institute of Technology Jodhpur in Computer Science and Engineering. During my doctoral studies, I had the opportunity to collaborate with the Computer Vision Center (CVC) in Barcelona, where I honed my expertise in mitigating biases within Language-Vision models. I later returned to CVC as a postdoctoral researcher, focusing on creating advanced tools for multimodal fake news verification.
What excites you about the field you’re working in?
In an era defined by the rapid spread of misinformation, the challenge of developing technologies that can cut through disinformation excites me. The real-world application of these tools, beyond controlled environments, is crucial. My focus is on building systems that are globally relevant—spanning languages, cultures, and media types—while also ensuring they are transparent and explainable. I’m driven by the idea that we can give people the tools to critically assess information rather than simply relying on algorithms to make decisions for them.
