Publications (22)
Fixing the train-test resolution discrepancy: FixEfficientNet
Hugo Touvron, Andrea Vedaldi, Matthijs Douze +1
This paper provides an extensive analysis of the performance of the EfficientNet image classifiers with several recent training procedures, in particular one that corrects the disc…
Are Large-scale Datasets Necessary for Self-Supervised Pre-training?
Alaaeldin El-Nouby, Gautier Izacard, Hugo Touvron +3
Pre-training models on large scale datasets, like ImageNet, is a standard practice in computer vision. This paradigm is especially effective for tasks with small training sets, for…
Three things everyone should know about Vision Transformers
Hugo Touvron, Matthieu Cord, Alaaeldin El-Nouby +2
After their initial success in natural language processing, transformer architectures have rapidly gained traction in computer vision, providing state-of-the-art results for tasks…
Llama 2: Open Foundation and Fine-Tuned Chat Models
Hugo Touvron, Louis Martin, Kevin Stone +65
In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our f…
LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference
Ben Graham, Alaaeldin El-Nouby, Hugo Touvron +4
We design a family of image classification architectures that optimize the trade-off between accuracy and efficiency in a high-speed regime. Our work exploits recent findings in at…
LLaMA: Open and Efficient Foundation Language Models
Hugo Touvron, Thibaut Lavril, Gautier Izacard +11
We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train…
ResMLP: Feedforward networks for image classification with data-efficient training
Hugo Touvron, Piotr Bojanowski, Mathilde Caron +8
We present ResMLP, an architecture built entirely upon multi-layer perceptrons for image classification. It is a simple residual network that alternates (i) a linear layer in which…
Emerging Properties in Self-Supervised Vision Transformers
Mathilde Caron, Hugo Touvron, Ishan Misra +4
In this paper, we question if self-supervised learning provides new properties to Vision Transformer (ViT) that stand out compared to convolutional networks (convnets). Beyond the…
Co-training Submodels for Visual Recognition
Hugo Touvron, Matthieu Cord, Maxime Oquab +3
We introduce submodel co-training, a regularization method related to co-training, self-distillation and stochastic depth. Given a neural network to be trained, for each sample we…
ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases
Stéphane d'Ascoli, Hugo Touvron, Matthew Leavitt +3
Convolutional architectures have proven extremely successful for vision tasks. Their hard inductive biases enable sample-efficient learning, but come at the cost of a potentially l…
Code Llama: Open Foundation Models for Code
Baptiste Rozière, Jonas Gehring, Fabian Gloeckle +23
We release Code Llama, a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for lar…
Powers of layers for image-to-image translation
Hugo Touvron, Matthijs Douze, Matthieu Cord +1
We propose a simple architecture to address unpaired image-to-image translation tasks: style or class transfer, denoising, deblurring, deblocking, etc. We start from an image autoe…
Automatic Data Curation for Self-Supervised Learning: A Clustering-Based Approach
Huy V. Vo, Vasil Khalidov, Timothée Darcet +10
Self-supervised features are the cornerstone of modern machine learning systems. They are typically pre-trained on data collections whose construction and curation typically requir…
Augmenting Convolutional networks with attention-based aggregation
Hugo Touvron, Matthieu Cord, Alaaeldin El-Nouby +4
We show how to augment any convolutional network with an attention-based global map to achieve non-local reasoning. We replace the final average pooling by an attention-based aggre…
Grafit: Learning fine-grained image representations with coarse labels
Hugo Touvron, Alexandre Sablayrolles, Matthijs Douze +2
This paper tackles the problem of learning a finer representation than the one provided by training labels. This enables fine-grained category retrieval of images in a collection a…
Going deeper with Image Transformers
Hugo Touvron, Matthieu Cord, Alexandre Sablayrolles +2
Transformers have been recently adapted for large scale image classification, achieving high scores shaking up the long supremacy of convolutional neural networks. However the opti…
The Llama 3 Herd of Models
Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri +556
Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models th…
XCiT: Cross-Covariance Image Transformers
Alaaeldin El-Nouby, Hugo Touvron, Mathilde Caron +8
Following their success in natural language processing, transformers have recently shown much promise for computer vision. The self-attention operation underlying transformers yiel…
Fixing the train-test resolution discrepancy
Hugo Touvron, Andrea Vedaldi, Matthijs Douze +1
Data-augmentation is key to the training of neural networks for image classification. This paper first shows that existing augmentations induce a significant discrepancy between th…
Training data-efficient image transformers & distillation through attention
Hugo Touvron, Matthieu Cord, Matthijs Douze +3
Recently, neural networks purely based on attention were shown to address image understanding tasks such as image classification. However, these visual transformers are pre-trained…
ResNet strikes back: An improved training procedure in timm
Ross Wightman, Hugo Touvron, Hervé Jégou
The influential Residual Networks designed by He et al. remain the gold-standard architecture in numerous scientific publications. They typically serve as the default architecture…
DeiT III: Revenge of the ViT
Hugo Touvron, Matthieu Cord, Hervé Jégou
A Vision Transformer (ViT) is a simple neural architecture amenable to serve several computer vision tasks. It has limited built-in architectural priors, in contrast to more recent…