Alaaeldin El-Nouby

I am a Staff Research Scientist at Meta Superintelligence Labs (FAIR) working on agentic coding post-training and previously LLM pre-training. Before that, I was part of the MLR team at Apple where I led research in scaling vision and multimodal pre-training.

I completed my PhD in Computer Science at Meta AI (FAIR) and École Normale Supérieure, advised by Ivan Laptev, Natalia Neverova, and Hervé Jégou. I have a MSc in Computer Engineering from University of Guelph, where I was advised by Dr. Graham Taylor. During that time, I was a student researcher at the Vector Institute.

Email: alaaelnouby-at-gmail.com / Google Scholar / Resume

News

Apr 2026 Meta announced Muse Spark, its latest frontier LLM. My team contributed to its agentic coding capabilities.
Jun 2025 Joined Meta Superintelligence Labs (FAIR).
Apr 2025 Scaling Laws for Native Multimodal Models was accepted at ICCV 2025 as an oral.
Feb 2025 Released FlexTok, accepted at ICML 2025.
Jan 2025 Parameters vs FLOPs was accepted at ICML 2025.
Nov 2024 Released AIMv2, accepted at CVPR 2025 as a highlight.
Jun 2024 Released DataComp-LM, accepted at NeurIPS 2024.
Jan 2024 Released Autoregressive Image Models (AIM), accepted at ICML 2024.
Aug 2023 Joined Apple MLR as a Research Scientist.
May 2023 Mark Zuckerberg announced our recent foundational multimodal model ImageBind.
May 2023 Released ImageBind, accepted at CVPR 2023.

Research

I'm interested in agentic coding post-training, LLM pre-training, multimodal vision language models, large-scale visual representation learning.

	Scaling Laws for Optimal Data Mixtures Mustafa Shukor, Louis Bethune, Dan Busbridge, David Grangier, Enrico Fini, Alaaeldin El-Nouby, Pierre Ablin NeurIPS 2025 paper
	Scaling Laws for Native Multimodal Models Mustafa Shukor, Enrico Fini, Victor Guilherme Turrisi da Costa, Matthieu Cord, Joshua Susskind, Alaaeldin El-Nouby ICCV 2025 Oral paper
	FlexTok: Resampling Images into 1D Token Sequences of Flexible Length Roman Bachmann, Jesse Allardice, David Mizrahi, Enrico Fini, Oguzhan Fatih Kar, Elmira Amirloo, Alaaeldin El-Nouby, Amir Zamir, Afshin Dehghan ICML 2025 paper code
	Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models Samira Abnar, Harshay Shah, Dan Busbridge, Alaaeldin El-Nouby, Josh Susskind, Vimal Thilak ICML 2025 paper
	Multimodal Autoregressive Pre-training of Large Vision Encoders Enrico Fini, Mustafa Shukor, Xiujun Li, Philipp Dufter, Michal Klein, David Haldimann, Sai Aitharaju, Victor Guilherme Turrisi da Costa, Louis Bethune, Zhe Gan, Alexander T Toshev, Marcin Eichner, Moin Nabi, Yinfei Yang, Joshua M. Susskind, Alaaeldin El-Nouby CVPR 2025 Highlight paper code
	DataComp-LM: In Search of the Next Generation of Training Sets for Language Models Jeffrey Li, Alex Fang, Georgios Smyrnis, Maor Ivgi, Matt Jordan, Vaishaal Shankar et al. NeurIPS 2024 paper code
	Scalable Pre-training of Large Autoregressive Image Models Alaaeldin El-Nouby, Michal Klein, Shuangfei Zhai, Miguel Angel Bautista, Alexander Toshev, Vaishaal Shankar, Joshua M. Susskind, Armand Joulin ICML 2024 paper code
	ImageBind: One Embedding Space To Bind Them All Rohit Girdhar, Alaaeldin El-Nouby, Zhuang Liu, Mannat Singh, Kalyan Vasudev Alwala, Armand Joulin, Ishan Misra CVPR 2023 Highlight paper code
	DINOv2: Learning Robust Visual Features without Supervision Maxime Oquab, Timothee Darcet, Theo Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Mahmoud Assran, Nicolas Ballas, Wojciech Galuba, Russell Howes, Po-Yao Huang, Shang-Wen Li, Ishan Misra, Michael Rabbat, Vasu Sharma, Gabriel Synnaeve, Hu Xu, Herve Jegou, Julien Mairal, Patrick Labatut, Armand Joulin, Piotr Bojanowski Preprint paper code
	Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models Matthew J. Muckley, Alaaeldin El-Nouby, Karen Ullrich, Herve Jegou, Jakob Verbeek ICML 2023 paper
	Image Compression with Product Quantized Masked Image Modeling Alaaeldin El-Nouby, Matthew J. Muckley, Karen Ullrich, Ivan Laptev, Jakob Verbeek, Hervé Jégou Transactions of Machine Learning Research (TMLR) paper
	OmniMAE: Single Model Masked Pretraining on Images and Videos Rohit Girdhar, Alaaeldin El-Nouby, Mannat Singh, Kalyan Vasudev Alwala, Armand Joulin, Ishan Misra* CVPR 2023 paper code
	Three things everyone should know about Vision Transformers Hugo Touvron, Matthieu Cord, Alaaeldin El-Nouby, Jakob Verbeek, Hervé Jégou ECCV 2022 paper code
	Are Large-scale Datasets Necessary for Self-Supervised Pre-training? Alaaeldin El-Nouby, Gautier Izacard, Hugo Touvron, Ivan Laptev, Hervé Jegou, Edouard Grave Under Review paper
	XCiT: Cross-Covariance Image Transformer Alaaeldin El-Nouby, Hugo Touvron, Mathilde Caron, Piotr Bojanowski, Matthijs Douze, Armand Joulin, Ivan Laptev, Natalia Neverova, Gabriel Synnaeve, Jakob Verbeek, Hervé Jegou NeurIPS 2021 paper video code
	ResMLP: Feedforward networks for image classification with data-efficient training Hugo Touvron, Piotr Bojanowski, Mathilde Caron, Matthieu Cord, Alaaeldin El-Nouby, Edouard Grave, Gautier Izacard, Armand Joulin, Gabriel Synnaeve, Jakob Verbeek, Hervé Jégou TPAMI paper code
	LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference Ben Graham, Alaaeldin El-Nouby, Hugo Touvron, Pierre Stock, Armand Joulin, Hervé Jégou, Matthijs Douze ICCV 2021 paper code
	Training Vision Transformers for Image Retrieval Alaaeldin El-Nouby, Natalia Neverova, Ivan Laptev, Hervé Jégou Preprint paper
	Skip-Clip: Self-Supervised Spatiotemporal Representation Learning by Future Clip Order Ranking Alaaeldin El-Nouby, Shuangfei Zhai, Graham W. Taylor, Joshua M. Susskind Holistic Video Understanding Workshop ICCV2019 (Best poster Award) paper poster bibtex
	Tell, Draw, and Repeat: Generating and Modifying Images Based on Continual Linguistic Instruction Alaaeldin El-Nouby, Shikhar Sharma, Hannes Schulz, Devon Hjelm, Layla El Asri, Samira Ebrahimi Kahou, Yoshua Bengio, Graham W.Taylor Proceedings of the 2019 IEEE International Conference on Computer Vision (ICCV) paper code poster blog bibtex
	Real-Time End-to-End Action Detection with Two-Stream Networks Alaaeldin El-Nouby, Graham W. Taylor 15th Conference on Computer and Robot Vision, CRV 2018 Oral paper bibtex
	Spatiotemporal Representation Learning For Human Action Recognition And Localization Alaaeldin El-Nouby paper bibtex

Invited Talks

L3D-IVU Workshop, CVPR’24 - Scalable Pre-training of Large Autoregressive Image Models
IMAGINE lab, École des Ponts ParisTech - Bringing the power of Transformers to Computer Vision
Vector Institute / University of Guelph - Masked Image Modeling for Visual Representation Learning
Transformers for Vision workshop, CVPR’22 - Are Large-scale Datasets Necessary for Self-Supervised Pre-training?
Max Planck Institute / Tübingen AI Center - Are Large-scale Datasets Necessary for Self-Supervised Pre-training?
Computer Vision Group (CVG), University of Bern - Are Large-scale Datasets Necessary for Self-Supervised Pre-training?
Large Scale Holistic Video Understanding workshop, CVPR’21 - Training Vision Transformers for Image Retrieval
KTH Royal Institute of Technology - Training Vision Transformers for Image Retrieval
Microsoft Research Montreal - Sequential Scene Understanding and Generation
DeepVision workshop, Simon Fraser University - Real-Time End-to-End Action Detection with Two-Stream Networks
Twenty Billion Neurons - Real-Time End-to-End Action Detection with Two-Stream Networks

Reviewing

CVPR’22-'25, ECCV’22-'24, NeurIPS’24, ICLR’24, ICCV’21, NeurIPS’22 SSL workshop, TPAMI (2021-Present).

Design and source code from Jon Barron's website

Alaaeldin El-Nouby

News

Research

Scaling Laws for Optimal Data Mixtures

Scaling Laws for Native Multimodal Models

FlexTok: Resampling Images into 1D Token Sequences of Flexible Length

Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models

Multimodal Autoregressive Pre-training of Large Vision Encoders

DataComp-LM: In Search of the Next Generation of Training Sets for Language Models

Scalable Pre-training of Large Autoregressive Image Models

ImageBind: One Embedding Space To Bind Them All

DINOv2: Learning Robust Visual Features without Supervision

Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models

Image Compression with Product Quantized Masked Image Modeling

OmniMAE: Single Model Masked Pretraining on Images and Videos

Three things everyone should know about Vision Transformers

Are Large-scale Datasets Necessary for Self-Supervised Pre-training?

XCiT: Cross-Covariance Image Transformer

ResMLP: Feedforward networks for image classification with data-efficient training

LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference

Training Vision Transformers for Image Retrieval

Skip-Clip: Self-Supervised Spatiotemporal Representation Learning by Future Clip Order Ranking

Tell, Draw, and Repeat: Generating and Modifying Images Based on Continual Linguistic Instruction

Real-Time End-to-End Action Detection with Two-Stream Networks

Spatiotemporal Representation Learning For Human Action Recognition And Localization

Invited Talks

Reviewing