Oscar Mañas
VISITING RESEARCHER | MONTREAL, CANADA
I am a visiting researcher at Meta FAIR, advised by Dr. Michal Drozdzal and Prof. Adriana Romero. I am also a PhD candidate at Mila and Université de Montréal, advised by Prof. Aishwarya Agrawal.
My research interests lie at the intersection of computer vision and natural language processing. I believe that, like humans (and other animals), AI systems should have a holistic understanding of the world around them. This means working with multiple sensory modalities, among which vision and language arise as particularly interesting. My work focuses on multimodal vision-language generative models, i.e. models capable of generating images and/or text conditioned on multimodal inputs.
Oscar's Work
Oscar's Publications
October 19, 2025
RESEARCH
NLP
Controlling Multimodal LLMs via Reward-guided Decoding
Aishwarya Agrawal, Adriana Romero Soriano, Koustuv Sinha, Michal Drozdzal, Oscar Mañas, Pierluca D'Oro
October 19, 2025
December 12, 2024
COMPUTER VISION
EvalGIM: A Library for Evaluating Generative Image Models
Melissa Hall, Abhishek Charnalia, Adriana Romero Soriano, Candace Ross, Carolina Braga, Jakob Verbeek, Karen Ullrich, Maeve Ryan, Mark Ibrahim, Marton Havasi, Michal Drozdzal, Mike Rabbat, Oscar Mañas, Pietro Astolfi, Reyhane Askari, Tariq Berrada Ifriqi, Yohann Benchetrit
December 12, 2024
June 05, 2024
CORE MACHINE LEARNING
An Introduction to Vision-Language Modeling
Anurag Ajay, Alexander C. Li, Suzanne Petryk, Zhiqiu Lin, Anas Mahmoud, Jun Chen, Mazda Moayeri, Aishwarya Agrawal, Florian Bordes, Adrien Bardes, Arjang Talattof, Asli Celikyilmaz, Bargav Jayaraman, Candace Ross, Chuan Guo, Diane Bouchacourt, Ellen Tan, Haider Al-Tahan, Hu Xu, Jonathan Lebensold, Kamalika Chaudhuri, Karen Ullrich, Karthik Padthe, Kate Saenko, Kushal Tirumala, Mark Ibrahim, Megan Richards, Melissa Hall, Oscar Mañas, Pietro Astolfi, Quentin Garrido, Reyhane Askari, Richard Pang, Rim Assouel, Samuel Lavoie, Srihari Jayakumar, Vasu Sharma, Vikas Chandra, Xilun Chen, Yunyang Xiong, Zechun Liu
June 05, 2024
Our approach
Latest news
Foundational models
Meta © 2026