Related papers: Decorrelation Speeds Up Vision Transformers

Learning transferable representations from unlabeled time series is crucial for improving performance in data-scarce classification. Existing self-supervised methods often operate at the point level and rely on unidirectional encoding,…

Audio classification and restoration are among major downstream tasks in audio signal processing. However, restoration derives less of a benefit from pretrained models compared to the overwhelming success of pretrained models in…

We introduce the concept of a Modular Autoencoder (MAE), capable of learning a set of diverse but complementary representations from unlabelled data, that can later be used for supervised tasks. The learning of the representations is…

Hybrid Mamba-Transformer networks have recently garnered broad attention. These networks can leverage the scalability of Transformers while capitalizing on Mamba's strengths in long-context modeling and computational efficiency. However,…

In this paper we tackle a fundamental question: "Can we train latent diffusion models together with the variational auto-encoder (VAE) tokenizer in an end-to-end manner?" Traditional deep-learning wisdom dictates that end-to-end training is…

Masked image modeling (MIM) has become a leading self-supervised learning strategy. MIMs such as Masked Autoencoder (MAE) learn strong representations by randomly masking input tokens for the encoder to process, with the decoder…

Masked autoencoders (MAEs) have established themselves as a powerful method for unsupervised pre-training for computer vision tasks. While vanilla MAEs put equal emphasis on reconstructing the individual parts of the image, we propose to…

In this paper, we investigate the adversarial robustness of vision transformers that are equipped with BERT pretraining (e.g., BEiT, MAE). A surprising observation is that MAE has significantly worse adversarial robustness than other BERT…

Recent works have shown that visual pretraining on egocentric datasets using masked autoencoders (MAE) can improve generalization for downstream robotics tasks. However, these approaches pretrain only on 2D images, while many robotics…

Deep supervision, which involves extra supervisions to the intermediate features of a neural network, was widely used in image classification in the early deep learning era since it significantly reduces the training difficulty and eases…

Current popular backbones in computer vision, such as Vision Transformers (ViT) and ResNets are trained to perceive the world from 2D images. However, to more effectively understand 3D structural priors in 2D backbones, we propose Mask3D to…

Text variational autoencoders (VAEs) are notorious for posterior collapse, a phenomenon where the model's decoder learns to ignore signals from the encoder. Because posterior collapse is known to be exacerbated by expressive decoders,…

As a class of fruitful approaches, diffusion probabilistic models (DPMs) have shown excellent advantages in high-resolution image reconstruction. On the other hand, masked autoencoders (MAEs), as popular self-supervised vision learners,…

Recent token reduction methods for Vision Transformers (ViTs) incorporate token merging, which measures the similarities between token embeddings and combines the most similar pairs. However, their merging policies are directly dependent on…

Prompt tuning (PT), as an emerging resource-efficient fine-tuning paradigm, has showcased remarkable effectiveness in improving the task-specific transferability of vision-language models. This paper delves into a previously overlooked…

Vision Transformer (ViT) has become one of the most popular neural architectures due to its great scalability, computational efficiency, and compelling performance in many vision tasks. However, ViT has shown inferior performance to…

Variational autoencoders (VAEs) typically encode images into a compact latent space, reducing computational cost but introducing an optimization dilemma: a higher-dimensional latent space improves reconstruction fidelity but often hampers…

Learning a robust video Variational Autoencoder (VAE) is essential for reducing video redundancy and facilitating efficient video generation. Directly applying image VAEs to individual frames in isolation can result in temporal…

The variational autoencoder (VAE) is a popular combination of deep latent variable model and accompanying variational learning technique. By using a neural inference network to approximate the model's posterior on latent variables, VAEs…

Denoising-based diffusion transformers, despite their strong generation performance, suffer from inefficient training convergence. Existing methods addressing this issue, such as REPA (relying on external representation encoders) or SRA…