Yann LeCun

42

Papers

3,872

Total Citations

Papers (42)

Efficient Object Localization Using Convolutional Networks

Computing the Stereo Matching Cost With a Convolutional Neural Network

Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs

Disentangling factors of variation in deep representation using adversarial training

NeurIPS 2016arXiv

Learning to Linearize Under Uncertainty

NeurIPS 2015arXiv

Navigation World Models

Layer by Layer: Uncovering Hidden Representations in Language Models

Transformers without Normalization

Scaling Language-Free Visual Representation Learning

Hierarchical World Models as Visual Whole-Body Humanoid Controllers

RoboPEPP: Vision-Based Robot Pose and Joint Angle Estimation through Embedding Predictive Pre-Training

Seq-VCR: Preventing Collapse in Intermediate Transformer Representations for Enhanced Reasoning

Character-level Convolutional Networks for Text Classification

Deep learning with Elastic Averaging SGD

NeurIPS 2015arXiv

How Learning by Reconstruction Produces Uninformative Features For Perception

Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation

MetaMorph: Multimodal Understanding and Generation via Instruction Tuning

The Entropy Enigma: Success and Failure of Entropy Minimization

Stochastic positional embeddings improve masked image modeling

A Closer Look at Spatiotemporal Convolutions for Action Recognition

Unsupervised Image Matching and Object Discovery as Optimization

Self-Supervised Learning From Images With a Joint-Embedding Predictive Architecture

Unsupervised Learning of Spatiotemporally Coherent Metrics

Predicting Deeper Into the Future of Semantic Segmentation

MDETR - Modulated Detection for End-to-End Multi-Modal Understanding

Decoupled Contrastive Learning

GLoMo: Unsupervised Learning of Transferable Relational Graphs

VICRegL: Self-Supervised Learning of Local Visual Features

projUNN: efficient method for training deep networks with unitary matrices

A Data-Augmentation Is Worth A Thousand Samples: Analytical Moments And Sampling-Free Training

Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods

Pre-Train Your Loss: Easy Bayesian Transfer Learning with Informative Priors

Coarse-to-Fine Vision-Language Pre-training with Fusion in the Backbone

The Effects of Regularization and Data Augmentation are Class Dependent

Self-Supervised Learning with Lie Symmetries for Partial Differential Equations

An Information Theory Perspective on Variance-Invariance-Covariance Regularization

Reverse Engineering Self-Supervised Learning

Binary embeddings with structured hashed projections

Recurrent Orthogonal Networks and Long-Memory Tasks

Tunable Efficient Unitary Neural Networks (EUNN) and their application to RNNs

Comparing Dynamics: Deep Neural Networks versus Glassy Systems

Adversarially Regularized Autoencoders