An Efficient Adaptive Compression Method for Human Perception and Machine Vision Tasks

While most existing neural image compression (NIC) and neural video compression (NVC) methodologies have achieved remarkable success, their optimization is primarily focused on human visual perception. However, with the rapid development of artificial intelligence, many images and videos will be used for various machine vision tasks. Consequently, such existing compression methodologies cannot achieve competitive performance in machine vision. In this work, we introduce an efficient adaptive compression (EAC) method tailored for both human perception and multiple machine vision tasks. Our method involves two key modules: 1), an adaptive compression mechanism, that adaptively selects several subsets from latent features to balance the optimizations for multiple machine vision tasks (e.g., segmentation, and detection) and human vision. 2), a task-specific adapter, that uses the parameter-efficient delta-tuning strategy to stimulate the comprehensive downstream analytical networks for specific machine vision tasks. By using the above two modules, we can optimize the bit-rate costs and improve machine vision performance. In general, our proposed EAC can seamlessly integrate with existing NIC (i.e., Ball\'e2018, and Cheng2020) and NVC (i.e., DVC, and FVC) methods. Extensive evaluation on various benchmark datasets (i.e., VOC2007, ILSVRC2012, VOC2012, COCO, UCF101, and DAVIS) shows that our method enhances performance for multiple machine vision tasks while maintaining the quality of human vision.

Exploring Optimal Latent Trajetory for Zero-shot Image Editing

Editability and fidelity are two essential demands for text-driven image editing, which expects that the editing area should align with the target prompt and the rest should remain unchanged separately. The current cutting-edge editing methods usually obey an "inversion-then-editing" pipeline, where the source image is first inverted to an approximate Gaussian noise ${z}_T$, based on which a sampling process is conducted using the target prompt. Nevertheless, we argue that it is not a good choice to use a near-Gaussian noise as a pivot for further editing since it almost lost all structure fidelity. We verify this by a pilot experiment, discovering that some intermediate-inverted latents can achieve a better trade-off between editability and fidelity than the fully-inverted ${z}_T$. Based on this, we propose a novel editing paradigm dubbed ZZEdit, which gentlely strengthens the target guidance on a sufficient-for-editing while structure-preserving latent. Specifically, we locate such an editing pivot by searching the first point on the inversion trajectory which has larger response levels toward the target prompt than the source one. Then, we propose a ZigZag process to perform mild target guiding on this pivot, which fulfills denoising and inversion iteratively, approaching the target while still holding fidelity. Afterwards, to achieve the same number of inversion and denoising steps, we perform a pure sampling process under the target prompt. Extensive experiments highlight the effectiveness of our ZZEdit in diverse image editing scenarios compared with the "inversion-then-editing" pipeline.

A Comprehensive Forecasting Framework based on Multi-Stage Hierarchical Forecasting Reconciliation and Adjustment

Ads demand forecasting for Walmart's ad products plays a critical role in enabling effective resource planning, allocation, and management of ads performance. In this paper, we introduce a comprehensive demand forecasting system that tackles hierarchical time series forecasting in business settings. Though traditional hierarchical reconciliation methods ensure forecasting coherence, they often trade off accuracy for coherence especially at lower levels and fail to capture the seasonality unique to each time-series in the hierarchy. Thus, we propose a novel framework "Multi-Stage Hierarchical Forecasting Reconciliation and Adjustment (Multi-Stage HiFoReAd)" to address the challenges of preserving seasonality, ensuring coherence, and improving accuracy. Our system first utilizes diverse models, ensembled through Bayesian Optimization (BO), achieving base forecasts. The generated base forecasts are then passed into the Multi-Stage HiFoReAd framework. The initial stage refines the hierarchy using Top-Down forecasts and "harmonic alignment." The second stage aligns the higher levels' forecasts using MinTrace algorithm, following which the last two levels undergo "harmonic alignment" and "stratified scaling", to eventually achieve accurate and coherent forecasts across the whole hierarchy. Our experiments on Walmart's internal Ads-demand dataset and 3 other public datasets, each with 4 hierarchical levels, demonstrate that the average Absolute Percentage Error from the cross-validation sets improve from 3% to 40% across levels against BO-ensemble of models (LGBM, MSTL+ETS, Prophet) as well as from 1.2% to 92.9% against State-Of-The-Art models. In addition, the forecasts at all hierarchical levels are proved to be coherent. The proposed framework has been deployed and leveraged by Walmart's ads, sales and operations teams to track future demands, make informed decisions and plan resources.

Empowering LLMs to Understand and Generate Complex Vector Graphics

The unprecedented advancements in Large Language Models (LLMs) have profoundly impacted natural language processing but have yet to fully embrace the realm of scalable vector graphics (SVG) generation. While LLMs encode partial knowledge of SVG data from web pages during training, recent findings suggest that semantically ambiguous and tokenized representations within LLMs may result in hallucinations in vector primitive predictions. Additionally, LLM training typically lacks modeling and understanding of the rendering sequence of vector paths, which can lead to occlusion between output vector primitives. In this paper, we present LLM4SVG, an initial yet substantial step toward bridging this gap by enabling LLMs to better understand and generate vector graphics. LLM4SVG facilitates a deeper understanding of SVG components through learnable semantic tokens, which precisely encode these tokens and their corresponding properties to generate semantically aligned SVG outputs. Using a series of learnable semantic tokens, a structured dataset for instruction following is developed to support comprehension and generation across two primary tasks. Our method introduces a modular architecture to existing large language models, integrating semantic tags, vector instruction encoders, fine-tuned commands, and powerful LLMs to tightly combine geometric, appearance, and language information. To overcome the scarcity of SVG-text instruction data, we developed an automated data generation pipeline that collected a massive dataset of more than 250k SVG data and 580k SVG-text instructions, which facilitated the adoption of the two-stage training strategy popular in LLM development. By exploring various training strategies, we developed LLM4SVG, which significantly moves beyond optimized rendering-based approaches and language-model-based baselines to achieve remarkable results in human evaluation tasks.

SVGFusion: Scalable Text-to-SVG Generation via Vector Space Diffusion

The generation of Scalable Vector Graphics (SVG) assets from textual data remains a significant challenge, largely due to the scarcity of high-quality vector datasets and the limitations in scalable vector representations required for modeling intricate graphic distributions. This work introduces SVGFusion, a Text-to-SVG model capable of scaling to real-world SVG data without reliance on a text-based discrete language model or prolonged SDS optimization. The essence of SVGFusion is to learn a continuous latent space for vector graphics with a popular Text-to-Image framework. Specifically, SVGFusion consists of two modules: a Vector-Pixel Fusion Variational Autoencoder (VP-VAE) and a Vector Space Diffusion Transformer (VS-DiT). VP-VAE takes both the SVGs and corresponding rasterizations as inputs and learns a continuous latent space, whereas VS-DiT learns to generate a latent code within this space based on the text prompt. Based on VP-VAE, a novel rendering sequence modeling strategy is proposed to enable the latent space to embed the knowledge of construction logics in SVGs. This empowers the model to achieve human-like design capabilities in vector graphics, while systematically preventing occlusion in complex graphic compositions. Moreover, our SVGFusion's ability can be continuously improved by leveraging the scalability of the VS-DiT by adding more VS-DiT blocks. A large-scale SVG dataset is collected to evaluate the effectiveness of our proposed method. Extensive experimentation has confirmed the superiority of our SVGFusion over existing SVG generation methods, achieving enhanced quality and generalizability, thereby establishing a novel framework for SVG content creation. Code, model, and data will be released at: \href{https://ximinng.github.io/SVGFusionProject/}{https://ximinng.github.io/SVGFusionProject/}

Electrically functionalized body surface for deep-tissue bioelectrical recording

Directly probing deep tissue activities from body surfaces offers a noninvasive approach to monitoring essential physiological processes1-3. However, this method is technically challenged by rapid signal attenuation toward the body surface and confounding motion artifacts4-6 primarily due to excessive contact impedance and mechanical mismatch with conventional electrodes. Herein, by formulating and directly spray coating biocompatible two-dimensional nanosheet ink onto the human body under ambient conditions, we create microscopically conformal and adaptive van der Waals thin films (VDWTFs) that seamlessly merge with non-Euclidean, hairy, and dynamically evolving body surfaces. Unlike traditional deposition methods, which often struggle with conformality and adaptability while retaining high electronic performance, this gentle process enables the formation of high-performance VDWTFs directly on the body surface under bio-friendly conditions, making it ideal for biological applications. This results in low-impedance electrically functionalized body surfaces (EFBS), enabling highly robust monitoring of biopotential and bioimpedance modulations associated with deep-tissue activities, such as blood circulation, muscle movements, and brain activities. Compared to commercial solutions, our VDWTF-EFBS exhibits nearly two-orders of magnitude lower contact impedance and substantially reduces the extrinsic motion artifacts, enabling reliable extraction of bioelectrical signals from irregular surfaces, such as unshaved human scalps. This advancement defines a technology for continuous, noninvasive monitoring of deep-tissue activities during routine body movements.

MoTe: Learning Motion-Text Diffusion Model for Multiple Generation Tasks

Recently, human motion analysis has experienced great improvement due to inspiring generative models such as the denoising diffusion model and large language model. While the existing approaches mainly focus on generating motions with textual descriptions and overlook the reciprocal task. In this paper, we present~\textbf{MoTe}, a unified multi-modal model that could handle diverse tasks by learning the marginal, conditional, and joint distributions of motion and text simultaneously. MoTe enables us to handle the paired text-motion generation, motion captioning, and text-driven motion generation by simply modifying the input context. Specifically, MoTe is composed of three components: Motion Encoder-Decoder (MED), Text Encoder-Decoder (TED), and Moti-on-Text Diffusion Model (MTDM). In particular, MED and TED are trained for extracting latent embeddings, and subsequently reconstructing the motion sequences and textual descriptions from the extracted embeddings, respectively. MTDM, on the other hand, performs an iterative denoising process on the input context to handle diverse tasks. Experimental results on the benchmark datasets demonstrate the superior performance of our proposed method on text-to-motion generation and competitive performance on motion captioning.

Individual Content and Motion Dynamics Preserved Pruning for Video Diffusion Models

The high computational cost and slow inference time are major obstacles to deploying the video diffusion model (VDM) in practical applications. To overcome this, we introduce a new Video Diffusion Model Compression approach using individual content and motion dynamics preserved pruning and consistency loss. First, we empirically observe that deeper VDM layers are crucial for maintaining the quality of \textbf{motion dynamics} e.g., coherence of the entire video, while shallower layers are more focused on \textbf{individual content} e.g., individual frames. Therefore, we prune redundant blocks from the shallower layers while preserving more of the deeper layers, resulting in a lightweight VDM variant called VDMini. Additionally, we propose an \textbf{Individual Content and Motion Dynamics (ICMD)} Consistency Loss to gain comparable generation performance as larger VDM, i.e., the teacher to VDMini i.e., the student. Particularly, we first use the Individual Content Distillation (ICD) Loss to ensure consistency in the features of each generated frame between the teacher and student models. Next, we introduce a Multi-frame Content Adversarial (MCA) Loss to enhance the motion dynamics across the generated video as a whole. This method significantly accelerates inference time while maintaining high-quality video generation. Extensive experiments demonstrate the effectiveness of our VDMini on two important video generation tasks, Text-to-Video (T2V) and Image-to-Video (I2V), where we respectively achieve an average 2.5 $\times$ and 1.4 $\times$ speed up for the I2V method SF-V and the T2V method T2V-Turbo-v2, while maintaining the quality of the generated videos on two benchmarks, i.e., UCF101 and VBench.

HEMGS: A Hybrid Entropy Model for 3D Gaussian Splatting Data Compression

Fast progress in 3D Gaussian Splatting (3DGS) has made 3D Gaussians popular for 3D modeling and image rendering, but this creates big challenges in data storage and transmission. To obtain a highly compact 3DGS representation, we propose a hybrid entropy model for Gaussian Splatting (HEMGS) data compression, which comprises two primary components, a hyperprior network and an autoregressive network. To effectively reduce structural redundancy across attributes, we apply a progressive coding algorithm to generate hyperprior features, in which we use previously compressed attributes and location as prior information. In particular, to better extract the location features from these compressed attributes, we adopt a domain-aware and instance-aware architecture to respectively capture domain-aware structural relations without additional storage costs and reveal scene-specific features through MLPs. Additionally, to reduce redundancy within each attribute, we leverage relationships between neighboring compressed elements within the attributes through an autoregressive network. Given its unique structure, we propose an adaptive context coding algorithm with flexible receptive fields to effectively capture adjacent compressed elements. Overall, we integrate our HEMGS into an end-to-end optimized 3DGS compression framework and the extensive experimental results on four benchmarks indicate that our method achieves about 40\% average reduction in size while maintaining the rendering quality over our baseline method and achieving state-of-the-art compression results.

An In-depth Investigation of Sparse Rate Reduction in Transformer-like Models

Deep neural networks have long been criticized for being black-box. To unveil the inner workings of modern neural architectures, a recent work \cite{yu2024white} proposed an information-theoretic objective function called Sparse Rate Reduction (SRR) and interpreted its unrolled optimization as a Transformer-like model called Coding Rate Reduction Transformer (CRATE). However, the focus of the study was primarily on the basic implementation, and whether this objective is optimized in practice and its causal relationship to generalization remain elusive. Going beyond this study, we derive different implementations by analyzing layer-wise behaviors of CRATE, both theoretically and empirically. To reveal the predictive power of SRR on generalization, we collect a set of model variants induced by varied implementations and hyperparameters and evaluate SRR as a complexity measure based on its correlation with generalization. Surprisingly, we find out that SRR has a positive correlation coefficient and outperforms other baseline measures, such as path-norm and sharpness-based ones. Furthermore, we show that generalization can be improved using SRR as regularization on benchmark image classification datasets. We hope this paper can shed light on leveraging SRR to design principled models and study their generalization ability.