Srija Mukhopadhyay

Srija Mukhopadhyay

Student (and) Researcher

International Institute of Information Technology Hyderabad

About

Hi, I'm Srija! I am a Dual Degree (B.Tech + MS) student at IIIT Hyderabad, advised by Prof. Manish Shrivastava.

I am fascinated by how agents learn and interact, and crucially, how we can get reliable signals to measure their progress. I am constantly expanding my research horizons: my current interests lie in Multi-Agent Systems, Long Context Reasoning, and Reliable Evaluations. I also have been interested in AI Safety and Security, Reinforcement Learning and Multimodality for a while now. My goal is to produce impactful, open-source research that pushes the boundaries of safe and reliable AI systems.

Over the last few years, I’ve had the opportunity to intern at Google (thrice!), collaborate with Adobe Research, and work with amazing mentors and collaborators. You can find more details in the publications section and my CV.

Feel free to reach out to me via email or any of the social links above! I am always open to discussions, collaborations and new opportunities.

Selected Publications

I have had the privilege of working with Prof. Dan Roth, Prof. Manish Shrivastava, Prof. Ponnurangam Kumaraguru and Dr. Aparna Garimella among many other amazing researchers. Under their guidance, my research has spanned developing benchmarks for Multimodal LLMs, evaluating privacy in personalized AI, and exploring fairness in Federated Learning.

2024

Unraveling the Truth: Do VLMs really Understand Charts?

S. Mukhopadhyay*, A. Qidwai*, A. Garimella, P. Ramu, V. Gupta, D. Roth

EMNLP 2024

2025

PrivacyBench: A Conversational Benchmark for Evaluating Privacy in Personalized AI

S. Mukhopadhyay, S. Reddy, S. Muthukumar, J. An, P. Kumaraguru

Preprint

See all publications →