I'm a PhD student at Mila and University of Montreal, advised by Aishwarya Agrawal. My research is supported by the Fonds de recherche du Québec – Nature et technologies (FRQNT).
I am inspired by the Era of Experience vision for scalable intelligence and excited about research in physical AI, particularly improved training and representation learning in generative (world) models. To build agents that can operate effectively in large, complex environments, we need better representations and more scalable policy-gradient algorithms for learning from experience.
I feel a strange despair watching AI progress (can't literally sleep some nights)—the world we built from the age of reason now threatened by reasoning machines, yet I can't ignore the overwhelming beauty and potential of intelligence abundance. The closest description I've found is losing my religion.
Work in progress
I am excited about two works in progress that offer elegant solutions to representation learning and generative modeling from a unified perspective. I am sharing the current results here because I do not have the compute to scale them right now, but I continue to make progress on them. The code will be shared once I have a clean version ready.
Selected publications
See Google Scholar for the full list. *denotes equal contribution.
One Flow Transformer for Imagination and Control. Rabiul Awal, Jinseong Jeong, Ankur Sikarwar, Parisa Kordjamshidi, Andrii Zadaianchuk, Sai Rajeswar, Paul Hongsuck Seo, Aishwarya Agrawal. Under review, 2026. paper / website
Self-supervised Dynamics Forcing in Diffusion World Models. Rabiul Awal. Under review, 2026. paper
Grounding Computer-Use Agents from Demonstrations. Aarash Feizi, Shravan Nayak, Xiangru Jian, Kevin Qinghong Lin, Kaixin Li, Rabiul Awal + 11 others. ICLR 2026. paper / website
The Promise of RL for Autoregressive Image Editing. Saba Ahmadi*, Rabiul Awal*, Ankur Sikarwar*, Amirhossein Kazemnejad*, Ge Ya Luo, Juan Rodriguez, Sai Rajeswar, Siva Reddy, Chris Pal, Benno Krojer, Aishwarya Agrawal. NeurIPS 2025. paper / twitter
Rendering-Aware RL for Vector Graphics Generation. Juan A. Rodriguez*, Haotian Zhang*, Abhay Puri, Aarash Feizi, Rishav Pramanik, Pascal Wichmann, Arnab Mondal, Mohammad Reza, Rabiul Awal + 6 others. NeurIPS 2025. paper / twitter
CTRL-O: Language-Controllable Object-Centric Representations. Aniket Rajiv Didolkar*, Andrii Zadaianchuk*, Rabiul Awal*, Maximilian Seitzer, Efstratios Gavves, Aishwarya Agrawal. CVPR 2025 – Spotlight at MAR Workshop @ CVPR'25. paper / website / twitter
Misc
Talks I put extra care into, and a bit of teaching/organizing on the side.