Aryan Khurana

I am a final year undergraduate at BITS Pilani, where I am majoring in Electrical and Electronics Engineering. Currently, I am a research intern at Inria Sophia Antipolis with the STARS team, advised by Dr. François Brémond, where I work on video generation and understanding.

Previously, I was a research intern at NTNU, advised by Prof. Kiran Raja, where I worked on generative models for synthetic face images and their applications in identity documents. I was also a research intern at the University of Edinburgh, advised by Prof. Vaishak Belle, where I worked on interpretability, fairness, and trustworthiness in large language models. I was also part of the MultiNet team at the Manifold Research Group.

I am broadly interested in large language and multimodal generative models, and interpretability. Feel free to reach out if you'd like to discuss research or just chat!

Email / CV / LinkedIn / Scholar / Github

Research

	Authority, Truth, and Citation Bias: A Large-Scale Multi-Domain Benchmark for Studying Epistemic Susceptibility in Large Language Models Aryan Khurana, Aravind Ramana RN, Dhruv Kumar EIML Workshop & Trustworthy AI for Good Workshop, ICML, 2026 paper
	Generating ICAO-Compliant Synthetic Face Images via Curriculum-Guided Diffusion Raghavendra Mudgalgundurao, Patrick Schuch, Aryan Khurana, Raghavendra Ramachandra, Kiran Raja ICPR, 2026 (Oral Presentation) paper
	Orthogonal Projection Steering for Bias Mitigation and ICAO Compliance in Diffusion Transformers Raghavendra Mudgalgundurao, Aryan Khurana, Patrick Schuch, Raghavendra Ramachandra, Kiran Raja SynRDinBAS Workshop, WACV, 2026 (Oral Presentation) paper
	Do Multimodal Foundation Models Truly Generalize? Exposing Failure Modes Across Perception, Reasoning, and Action Pranav Guruprasad, Sudipta Chowdhury, Harsh Sikka, Mridul Sharma, Hong Lu, Sean Rivera, Hangliang Ren, Aryan Khurana, Yangyue Wang MMFM5 Workshop, CVPR, 2026 paper

Website template from Jon Barron.