Search arXiv⌕ Search

arXiv subjects

Yue Wu

Publications and source records attributed to Yue Wu.

At least 235 records · Page 13Linked to original sources
arXiv2024

PIXART-δ: Fast and Controllable Image Generation with Latent Consistency Models

This technical report introduces PIXART-δ, a text-to-image synthesis framework that integrates the Latent Consistency Model (LCM) and ControlNet into the advanced PIXART-α model. PIXART-α is recognized for its ability to generate high-quality images of 1024px resolution through a remarkably efficient training process. The integration of LCM in PIXART-δ significantly accelerates the inference speed, enabling the production of high-quality images in just 2-4 steps. Notably, PIXART-δ achieves a breakthrough 0.5 seconds for generating 1024x1024 pixel images, marking a 7x improvement over the PIXART-α. Additionally, PIXART-δ is designed to be efficiently trainable on 32GB V100 GPUs within a single day. With its 8-bit inference capability (von Platen et al., 2023), PIXART-δ can synthesize 1024px images within 8GB GPU memory constraints, greatly enhancing its usability and accessibility. Furthermore, incorporating a ControlNet-like module enables fine-grained control over text-to-image diffusion models. We introduce a novel ControlNet-Transformer architecture, specifically tailored for Transformers, achieving explicit controllability alongside high-quality image generation. As a state-of-the-art, open-source image generation model, PIXART-δ offers a promising alternative to the Stable Diffusion family of models, contributing significantly to text-to-image synthesis.

cs.CV↗
arXiv2023

PixArt-$α$: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis

The most advanced text-to-image (T2I) models require significant training costs (e.g., millions of GPU hours), seriously hindering the fundamental innovation for the AIGC community while increasing CO2 emissions. This paper introduces PIXART-$α$, a Transformer-based T2I diffusion model whose image generation quality is competitive with state-of-the-art image generators (e.g., Imagen, SDXL, and even Midjourney), reaching near-commercial application standards. Additionally, it supports high-resolution image synthesis up to 1024px resolution with low training cost, as shown in Figure 1 and 2. To achieve this goal, three core designs are proposed: (1) Training strategy decomposition: We devise three distinct training steps that separately optimize pixel dependency, text-image alignment, and image aesthetic quality; (2) Efficient T2I Transformer: We incorporate cross-attention modules into Diffusion Transformer (DiT) to inject text conditions and streamline the computation-intensive class-condition branch; (3) High-informative data: We emphasize the significance of concept density in text-image pairs and leverage a large Vision-Language model to auto-label dense pseudo-captions to assist text-image alignment learning. As a result, PIXART-$α$'s training speed markedly surpasses existing large-scale T2I models, e.g., PIXART-$α$ only takes 10.8% of Stable Diffusion v1.5's training time (675 vs. 6,250 A100 GPU days), saving nearly \$300,000 (\$26,000 vs. \$320,000) and reducing 90% CO2 emissions. Moreover, compared with a larger SOTA model, RAPHAEL, our training cost is merely 1%. Extensive experiments demonstrate that PIXART-$α$ excels in image quality, artistry, and semantic control. We hope PIXART-$α$ will provide new insights to the AIGC community and startups to accelerate building their own high-quality yet low-cost generative models from scratch.

cs.CV↗
arXiv2023

Targeted materials discovery using Bayesian algorithm execution

Rapid discovery and synthesis of new materials requires intelligent data acquisition strategies to navigate large design spaces. A popular strategy is Bayesian optimization, which aims to find candidates that maximize material properties; however, materials design often requires finding specific subsets of the design space which meet more complex or specialized goals. We present a framework that captures experimental goals through straightforward user-defined filtering algorithms. These algorithms are automatically translated into one of three intelligent, parameter-free, sequential data acquisition strategies (SwitchBAX, InfoBAX, and MeanBAX). Our framework is tailored for typical discrete search spaces involving multiple measured physical properties and short time-horizon decision making. We evaluate this approach on datasets for TiO$_2$ nanoparticle synthesis and magnetic materials characterization, and show that our methods are significantly more efficient than state-of-the-art approaches.

cond-mat.mtrl-sci↗
arXiv2023

A ternary mixture model with dynamic boundary conditions

The influence of short-range interactions between a multi-phase, multi-component mixture and a solid wall in confined geometries is crucial in life sciences and engineering. In this work, we extend the Cahn-Hilliard model with dynamic boundary conditions from a binary to a ternary mixture, employing the Onsager principle that accounts for cross-coupling between forces and fluxes in both bulk and surface. Moreover, we have developed a linear, second-order, and unconditionally energy stable numerical scheme for solving the governing equations, utilizing the Invariant Energy Quadratization (IEQ) method. This efficient solver allows us to explore the impacts of wall-mixture interactions and dynamic boundary conditions on phenomena like spontaneous phase separation, coarsening processes, and the wettability of droplets on surfaces. We observe that wall-mixture interactions influence not only surface phenomena, such as droplet contact angles, but also patterns deep within the bulk. Additionally, the relaxation rates control the droplet spreading on surfaces. Furthermore, the cross-coupling relaxation rates in the bulk significantly affect coarsening patterns. Our work establishes a comprehensive framework for studying multi-component mixtures in confined geometries.

physics.flu-dyn↗
arXiv2023

Neural Gaussian Similarity Modeling for Differential Graph Structure Learning

Graph Structure Learning (GSL) has demonstrated considerable potential in the analysis of graph-unknown non-Euclidean data across a wide range of domains. However, constructing an end-to-end graph structure learning model poses a challenge due to the impediment of gradient flow caused by the nearest neighbor sampling strategy. In this paper, we construct a differential graph structure learning model by replacing the non-differentiable nearest neighbor sampling with a differentiable sampling using the reparameterization trick. Under this framework, we argue that the act of sampling \mbox{nearest} neighbors may not invariably be essential, particularly in instances where node features exhibit a significant degree of similarity. To alleviate this issue, the bell-shaped Gaussian Similarity (GauSim) modeling is proposed to sample non-nearest neighbors. To adaptively model the similarity, we further propose Neural Gaussian Similarity (NeuralGauSim) with learnable parameters featuring flexible sampling behaviors. In addition, we develop a scalable method by transferring the large-scale graph to the transition graph to significantly reduce the complexity. Experimental results demonstrate the effectiveness of the proposed methods.

cs.LG↗
arXiv2023

Fast Training of Diffusion Transformer with Extreme Masking for 3D Point Clouds Generation

Diffusion Transformers have recently shown remarkable effectiveness in generating high-quality 3D point clouds. However, training voxel-based diffusion models for high-resolution 3D voxels remains prohibitively expensive due to the cubic complexity of attention operators, which arises from the additional dimension of voxels. Motivated by the inherent redundancy of 3D compared to 2D, we propose FastDiT-3D, a novel masked diffusion transformer tailored for efficient 3D point cloud generation, which greatly reduces training costs. Specifically, we draw inspiration from masked autoencoders to dynamically operate the denoising process on masked voxelized point clouds. We also propose a novel voxel-aware masking strategy to adaptively aggregate background/foreground information from voxelized point clouds. Our method achieves state-of-the-art performance with an extreme masking ratio of nearly 99%. Moreover, to improve multi-category 3D generation, we introduce Mixture-of-Expert (MoE) in 3D diffusion model. Each category can learn a distinct diffusion path with different experts, relieving gradient conflict. Experimental results on the ShapeNet dataset demonstrate that our method achieves state-of-the-art high-fidelity and diverse 3D point cloud generation performance. Our FastDiT-3D improves 1-Nearest Neighbor Accuracy and Coverage metrics when generating 128-resolution voxel point clouds, using only 6.5% of the original training cost.

cs.CV↗
arXiv2023

SPRING: Studying the Paper and Reasoning to Play Games

Open-world survival games pose significant challenges for AI algorithms due to their multi-tasking, deep exploration, and goal prioritization requirements. Despite reinforcement learning (RL) being popular for solving games, its high sample complexity limits its effectiveness in complex open-world games like Crafter or Minecraft. We propose a novel approach, SPRING, to read the game's original academic paper and use the knowledge learned to reason and play the game through a large language model (LLM). Prompted with the LaTeX source as game context and a description of the agent's current observation, our SPRING framework employs a directed acyclic graph (DAG) with game-related questions as nodes and dependencies as edges. We identify the optimal action to take in the environment by traversing the DAG and calculating LLM responses for each node in topological order, with the LLM's answer to final node directly translating to environment actions. In our experiments, we study the quality of in-context "reasoning" induced by different forms of prompts under the setting of the Crafter open-world environment. Our experiments suggest that LLMs, when prompted with consistent chain-of-thought, have great potential in completing sophisticated high-level trajectories. Quantitatively, SPRING with GPT-4 outperforms all state-of-the-art RL baselines, trained for 1M steps, without any training. Finally, we show the potential of games as a test bed for LLMs.

cs.AI↗
arXiv2023

Order-One Convergence of the Backward Euler Method for Random Periodic Solutions of Semilinear SDEs

In this paper, we revisit the backward Euler method for numerical approximations of random periodic solutions of semilinear SDEs with additive noise. Improved $L^{p}$-estimates of the random periodic solutions of the considered SDEs are obtained under a more relaxed condition compared to literature. The backward Euler scheme is proved to converge with an order one in the mean square sense, which also improves the existing order-half convergence. Numerical examples are presented to verify our theoretical analysis.

math.PR↗
arXiv2023

PCRDiffusion: Diffusion Probabilistic Models for Point Cloud Registration

We propose a new framework that formulates point cloud registration as a denoising diffusion process from noisy transformation to object transformation. During training stage, object transformation diffuses from ground-truth transformation to random distribution, and the model learns to reverse this noising process. In sampling stage, the model refines randomly generated transformation to the output result in a progressive way. We derive the variational bound in closed form for training and provide implementations of the model. Our work provides the following crucial findings: (i) In contrast to most existing methods, our framework, Diffusion Probabilistic Models for Point Cloud Registration (PCRDiffusion) does not require repeatedly update source point cloud to refine the predicted transformation. (ii) Point cloud registration, one of the representative discriminative tasks, can be solved by a generative way and the unified probabilistic formulation. Finally, we discuss and provide an outlook on the application of diffusion model in different scenarios for point cloud registration. Experimental results demonstrate that our model achieves competitive performance in point cloud registration. In correspondence-free and correspondence-based scenarios, PCRDifussion can both achieve exceeding 50\% performance improvements.

cs.CV↗
arXiv2023

M3SOT: Multi-frame, Multi-field, Multi-space 3D Single Object Tracking

3D Single Object Tracking (SOT) stands a forefront task of computer vision, proving essential for applications like autonomous driving. Sparse and occluded data in scene point clouds introduce variations in the appearance of tracked objects, adding complexity to the task. In this research, we unveil M3SOT, a novel 3D SOT framework, which synergizes multiple input frames (template sets), multiple receptive fields (continuous contexts), and multiple solution spaces (distinct tasks) in ONE model. Remarkably, M3SOT pioneers in modeling temporality, contexts, and tasks directly from point clouds, revisiting a perspective on the key factors influencing SOT. To this end, we design a transformer-based network centered on point cloud targets in the search area, aggregating diverse contextual representations and propagating target cues by employing historical frames. As M3SOT spans varied processing perspectives, we've streamlined the network-trimming its depth and optimizing its structure-to ensure a lightweight and efficient deployment for SOT applications. We posit that, backed by practical construction, M3SOT sidesteps the need for complex frameworks and auxiliary components to deliver sterling results. Extensive experiments on benchmarks such as KITTI, nuScenes, and Waymo Open Dataset demonstrate that M3SOT achieves state-of-the-art performance at 38 FPS. Our code and models are available at https://github.com/ywu0912/TeamCode.git.

cs.CV↗
arXiv2023

Drag-A-Video: Non-rigid Video Editing with Point-based Interaction

Video editing is a challenging task that requires manipulating videos on both the spatial and temporal dimensions. Existing methods for video editing mainly focus on changing the appearance or style of the objects in the video, while keeping their structures unchanged. However, there is no existing method that allows users to interactively ``drag'' any points of instances on the first frame to precisely reach the target points with other frames consistently deformed. In this paper, we propose a new diffusion-based method for interactive point-based video manipulation, called Drag-A-Video. Our method allows users to click pairs of handle points and target points as well as masks on the first frame of an input video. Then, our method transforms the inputs into point sets and propagates these sets across frames. To precisely modify the contents of the video, we employ a new video-level motion supervision to update the features of the video and introduce the latent offsets to achieve this update at multiple denoising timesteps. We propose a temporal-consistent point tracking module to coordinate the movement of the points in the handle point sets. We demonstrate the effectiveness and flexibility of our method on various videos. The website of our work is available here: https://drag-a-video.github.io/.

cs.CV↗
arXiv2023

Precision Agriculture: Crop Mapping using Machine Learning and Sentinel-2 Satellite Imagery

Food security has grown in significance due to the changing climate and its warming effects. To support the rising demand for agricultural products and to minimize the negative impact of climate change and mass cultivation, precision agriculture has become increasingly important for crop cultivation. This study employs deep learning and pixel-based machine learning methods to accurately segment lavender fields for precision agriculture, utilizing various spectral band combinations extracted from Sentinel-2 satellite imagery. Our fine-tuned final model, a U-Net architecture, can achieve a Dice coefficient of 0.8324. Additionally, our investigation highlights the unexpected efficacy of the pixel-based method and the RGB spectral band combination in this task.

cs.CV↗
arXiv2023

Open-Ended Instructable Embodied Agents with Memory-Augmented Large Language Models

Pre-trained and frozen large language models (LLMs) can effectively map simple scene rearrangement instructions to programs over a robot's visuomotor functions through appropriate few-shot example prompting. To parse open-domain natural language and adapt to a user's idiosyncratic procedures, not known during prompt engineering time, fixed prompts fall short. In this paper, we introduce HELPER, an embodied agent equipped with an external memory of language-program pairs that parses free-form human-robot dialogue into action programs through retrieval-augmented LLM prompting: relevant memories are retrieved based on the current dialogue, instruction, correction, or VLM description, and used as in-context prompt examples for LLM querying. The memory is expanded during deployment to include pairs of user's language and action plans, to assist future inferences and personalize them to the user's language and routines. HELPER sets a new state-of-the-art in the TEACh benchmark in both Execution from Dialog History (EDH) and Trajectory from Dialogue (TfD), with a 1.7x improvement over the previous state-of-the-art for TfD. Our models, code, and video results can be found in our project's website: https://helper-agent-llm.github.io.

cs.AI↗
arXiv2023

A high-order local discontinuous Galerkin method for the $p$-Laplace equation

We study the high-order local discontinuous Galerkin (LDG) method for the $p$-Laplace equation. We reformulate our spatial discretization as an equivalent convex minimization problem and use a preconditioned gradient descent method as the nonlinear solver. For the first time, a weighted preconditioner that provides $hk$-independent convergence is applied in the LDG setting. For polynomial order $k \geqslant 1$, we rigorously establish the solvability of our scheme and provide a priori error estimates in a mesh-dependent energy norm. Our error estimates are under a different and non-equivalent distance from existing LDG results. For arbitrarily high-order polynomials under the assumption that the exact solution has enough regularity, the error estimates demonstrate the potential for high-order accuracy. Our numerical results exhibit the desired convergence speed facilitated by the preconditioner, and we observe best convergence rates in gradient variables in alignment with linear LDG, and optimal rates in the primal variable when $1 < p \leqslant 2$.

math.NA↗
arXiv2023

NODE-ImgNet: a PDE-informed effective and robust model for image denoising

Inspired by the traditional partial differential equation (PDE) approach for image denoising, we propose a novel neural network architecture, referred as NODE-ImgNet, that combines neural ordinary differential equations (NODEs) with convolutional neural network (CNN) blocks. NODE-ImgNet is intrinsically a PDE model, where the dynamic system is learned implicitly without the explicit specification of the PDE. This naturally circumvents the typical issues associated with introducing artifacts during the learning process. By invoking such a NODE structure, which can also be viewed as a continuous variant of a residual network (ResNet) and inherits its advantage in image denoising, our model achieves enhanced accuracy and parameter efficiency. In particular, our model exhibits consistent effectiveness in different scenarios, including denoising gray and color images perturbed by Gaussian noise, as well as real-noisy images, and demonstrates superiority in learning from small image datasets.

eess.IV↗
arXiv2023

Energy cascade in the Garrett-Munk spectrum of internal gravity waves

We study the spectral energy transfer due to wave-triad interactions in the Garrett-Munk spectrum of internal gravity waves (IGWs) based on a numerical evaluation of the collision integral in the wave kinetic equation. Our numerical evaluation builds on the reduction of the collision integral on the resonant manifold for a horizontally isotropic spectrum. We directly evaluate the downscale energy flux available for ocean mixing, whose value is in close agreement with the empirical finescale parameterization. We further decompose the energy transfer into contributions from different mechanisms, including local interactions and three types of nonlocal interactions, namely parametric subharmonic instability (PSI), elastic scattering (ES) and induced diffusion (ID). Through analysis on the role of each type of interaction, we resolve two long-standing paradoxes regarding the mechanism for forward cascade in frequency and zero ID flux for GM76 spectrum. In addition, our analysis estimates the contribution of each mechanism to the energy transfer in each spectral direction, and reveals new understanding of the importance of local interactions and ES in the energy transfer.

physics.flu-dyn↗
arXiv2023

Generalized Permutation Framework for Testing Model Variable Significance

A common problem in machine learning is determining if a variable significantly contributes to a model's prediction performance. This problem is aggravated for datasets, such as gene expression datasets, that suffer the worst case of dimensionality: a low number of observations along with a high number of possible explanatory variables. In such scenarios, traditional methods for testing variable statistical significance or constructing variable confidence intervals do not apply. To address these problems, we developed a novel permutation framework for testing the significance of variables in supervised models. Our permutation framework has three main advantages. First, it is non-parametric and does not rely on distributional assumptions or asymptotic results. Second, it not only ranks model variables in terms of relative importance, but also tests for statistical significance of each variable. Third, it can test for the significance of the interaction between model variables. We applied this permutation framework to multi-class classification of the Iris flower dataset and of brain regions in RNA expression data, and using this framework showed variable-level statistical significance and interactions.

stat.ME↗
arXiv2023

Observations of a black hole X-ray binary indicate formation of a magnetically arrested disk

Accretion of material onto a black hole drags any magnetic fields present inwards, increasing their strength. Theory predicts that sufficiently strong magnetic fields can halt the accretion flow, producing a magnetically arrested disk (MAD). We analyze archival multi-wavelength observations of an outburst from the black hole X-ray binary MAXI J1820+070 in 2018. The radio and optical fluxes are delayed by about 8 and 17 days respectively, compared to the X-ray flux. We interpret this as evidence for the formation of a MAD. In this scenario, the magnetic field is amplified by an expanding corona, forming a MAD around the time of the radio peak. The optical delay is then due to thermal viscous instability in the outer disk.

astro-ph.HE↗