Related papers: Draft-and-Revise: Effective Image Generation with …

While recent Large Vision-Language Models (LVLMs) exhibit strong multimodal reasoning abilities, they often produce ungrounded or hallucinated responses because they rely too heavily on linguistic priors instead of visual evidence. This…

Masked generative models (MGMs) have emerged as a powerful framework for image synthesis, combining parallel decoding with strong bidirectional context modeling. However, generating high-quality samples typically requires many iterative…

Two fundamental challenges face generative models in engineering applications: the acquisition of high-performing, diverse datasets, and the adherence to precise constraints in generated designs. We propose a novel approach combining…

The use of coarse-grained layouts for controllable synthesis of complex scene images via deep generative models has recently gained popularity. However, results of current approaches still fall short of their promise of high-resolution…

Generative models have made significant progress in the tasks of modeling complex data distributions such as natural images. The introduction of Generative Adversarial Networks (GANs) and auto-encoders lead to the possibility of training on…

We propose a new paradigm to automatically generate training data with accurate labels at scale using the text-to-image synthesis frameworks (e.g., DALL-E, Stable Diffusion, etc.). The proposed approach1 decouples training data generation…

This thesis is a proof of concept for the potential of Variational Auto-Encoder (VAE) on representation learning of real-world Knowledge Graphs (KG). Inspired by successful approaches to the generation of molecular graphs, we evaluate the…

The ability to accurately model random fields plays a critical role in science and engineering for problems involving uncertain, spatially-varying quantities such as heterogeneous material properties and turbulent flows. Deep generative…

The current conditional autoregressive image generation methods have shown promising results, yet their potential remains largely unexplored in the practical unsupervised image translation domain, which operates without explicit…

Removing modeling constraints and unifying architectures across domains has been a key driver of the recent progress in training large multimodal models. However, most of these models still rely on many separately trained components such as…

Most 3D scene generation methods are limited to only generating object bounding box parameters while newer diffusion methods also generate class labels and latent features. Using object size or latent feature, they then retrieve objects…

This paper addresses the problem of lossy image compression, a fundamental problem in image processing and information theory that is involved in many real-world applications. We start by reviewing the framework of variational autoencoders…

Enabling image generation models to be spatially controlled is an important area of research, empowering users to better generate images according to their own fine-grained specifications via e.g. edge maps, poses. Although this task has…

In this work, we first revisit the sampling issues in current autoregressive (AR) image generation models and identify that image tokens, unlike text tokens, exhibit lower information density and non-uniform spatial distribution.…

Image Super-Resolution is a fundamental problem in computer vision with broad applications spacing from medical imaging to satellite analysis. The ability to reconstruct high-resolution images from low-resolution inputs is crucial for…

Current state-of-the-art generative approaches frequently rely on a two-stage training procedure, where an autoencoder (often a VAE) first performs dimensionality reduction, followed by training a generative model on the learned latent…

Objective image quality evaluation is a challenging task, which aims to measure the quality of a given image automatically. According to the availability of the reference images, there are Full-Reference and No-Reference IQA tasks,…

Vector quantized diffusion (VQ-Diffusion) is a powerful generative model for text-to-image synthesis, but sometimes can still generate low-quality samples or weakly correlated images with text input. We find these issues are mainly due to…

The Residual Quantization (RQ) framework is revisited where the quantization distortion is being successively reduced in multi-layers. Inspired by the reverse-water-filling paradigm in rate-distortion theory, an efficient regularization on…

Learning discrete representations with vector quantization (VQ) has emerged as a powerful approach in various generative models. However, most VQ-based models rely on a single, fixed-rate codebook, requiring extensive retraining for new…