I am an Assistant Professor at the Machine Learning department of the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). My research focuses on areas like LLM reasoning, GFlowNets, uncertainty estimation, reinforcement learning sample complexity, and more broadly, on designing better "AI for science" tools.

Research Topics

AI for Science Tools

LLM Reasoning

Curriculum Learning

Uncertainty Estimation

Sample Efficient RL

Bayesian Optimization

GFlowNets

Probabilistic Modeling

- Prospective graduate students are welcome to apply to MBZUAI and mention me as their supervisors after reaching out by email, mentioning their research experience and interests.
- Prospective visiting students with research experience, and interests aligned with mine are welcome to apply via email. I will not be responding to LLM-generated emails / research statements.

My Contributions +

I build more capable and reliable AI. My research focuses on teaching machines to reason, plan, and generate diverse solutions under uncertainty.

My work is centered on three key areas:

1. Developing Novel Methods: I am a core contributor to Generative Flow Networks (GFlowNets), a modern framework for structured probabilistic sampling. I also design new techniques for crucial problems like uncertainty estimation (DEUP) and efficient curriculum learning.
2. Advancing Large Language Models: I apply these methods to improve the reasoning and alignment of LLMs, with projects focused on enhancing multi-step logic and fine-tuning models with human preferences.
3. Building Tools for the Community: I believe in advancing science through shared tools. I created the torchgfn library for GFlowNets and have co-led the development of benchmarks like BabyAI, FinChain, and LLM-BabyBench.

Ultimately, I aim to create AI systems that are robust, creative, and trustworthy enough for real-world deployment.

News

• September 2025 - I gave an invited talk titled "Teaching LLMs to Reason Diversely: Process Reward Models and Step-Level GFlowNets for Mathematical Problem Solving" at Sorbonne University (Abu Dhabi)
• September 2025 - I will be serving as a co-Program chair at the PhD symposium of the Web Conference 2026
• May 2025 - I co-organized the International Symposium on Trustworthy Foundation Models at MBZUAI.
• May 2025 - I gave a tutorial titled "Reinforcement Learning through a probabilistic lens" (slides) at the METIS Spring School.
• May 2025 - I served as a co-Program Chair at the NETYS 2025 conference in Rabat.
• May 2025 - I gave a talk titled "Advancing Scientific Discovery with Generative AI" at the SKEMA Center for Artificial Intelligence.
• February 2025 - I gave a talk titled "GFlowNets: An Introduction and Recent Advances" at the MBZUAI-Berkeley Joint Workshop on Emerging Directions of Machine Learning.
• February 2025 - I gave a talk titled "GFlowNets: A Novel Framework for Diverse Generation in Combinatorial and Continuous Spaces" (slides) at the MBZUAI Paris Workshop 2025.
• December 2024 - I gave a keynote talk titled "Advancing the Fourth Paradigm: Machine Learning Frameworks for Experimental Science" at the the MoroccoAI 2024 conference.
• November 2024 - I attended the CLIMB workshop, and the Domain adaption and related areas workshop at the Simon's institute, in University of California, Berkeley.
• September 2024 - I joined MBZUAI as an Assistant Professor in the Machine Learning Department.

Ultimately, I am driven by a deep interest in understanding and defining intelligence, whether it's in animals or artificial systems. While I'm excited about the potential of AI, I'm also mindful of its limitations and potential pitfalls – I'm definitely not a technosolutionist.

Before my current role, I worked at TII as a Senior Researcher in 2024. I obtained my PhD in 2023 from Mila and UdeM, under the supervision of Yoshua Bengio. During my PhD, I also interned at Google Brain in Paris in 2020, focusing on hierarchical reinforcement learning. My academic background includes studies in applied mathematics at Ecole Polytechnique and statistical learning at Ecole Normale Supérieure Paris-Saclay. I've also explored research in game theory and operations research at IBM Research Singapore and gained industry experience as a data scientist at Booking.com in Amsterdam.