arXiv subjects

Tao Huang

Publications and source records attributed to Tao Huang.

At least 145 records · Page 8Linked to original sources

arXiv2025

Deep LG-Track: An Enhanced Localization-Confidence-Guided Multi-Object Tracker

Multi-object tracking plays a crucial role in various applications, such as autonomous driving and security surveillance. This study introduces Deep LG-Track, a novel multi-object tracker that incorporates three key enhancements to improve the tracking accuracy and robustness. First, an adaptive Kalman filter is developed to dynamically update the covariance of measurement noise based on detection confidence and trajectory disappearance. Second, a novel cost matrix is formulated to adaptively fuse motion and appearance information, leveraging localization confidence and detection confidence as weighting factors. Third, a dynamic appearance feature updating strategy is introduced, adjusting the relative weighting of historical and current appearance features based on appearance clarity and localization accuracy. Comprehensive evaluations on the MOT17 and MOT20 datasets demonstrate that the proposed Deep LG-Track consistently outperforms state-of-the-art trackers across multiple performance metrics, highlighting its effectiveness in multi-object tracking tasks.

cs.CV↗

arXiv2025

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders

Visual encoders are fundamental components in vision-language models (VLMs), each showcasing unique strengths derived from various pre-trained visual foundation models. To leverage the various capabilities of these encoders, recent studies incorporate multiple encoders within a single VLM, leading to a considerable increase in computational cost. In this paper, we present Mixture-of-Visual-Encoder Knowledge Distillation (MoVE-KD), a novel framework that distills the unique proficiencies of multiple vision encoders into a single, efficient encoder model. Specifically, to mitigate conflicts and retain the unique characteristics of each teacher encoder, we employ low-rank adaptation (LoRA) and mixture-of-experts (MoEs) to selectively activate specialized knowledge based on input features, enhancing both adaptability and efficiency. To regularize the KD process and enhance performance, we propose an attention-based distillation strategy that adaptively weighs the different encoders and emphasizes valuable visual tokens, reducing the burden of replicating comprehensive but distinct features from multiple teachers. Comprehensive experiments on popular VLMs, such as LLaVA and LLaVA-NeXT, validate the effectiveness of our method. Our code is available at: https://github.com/hey-cjj/MoVE-KD.

cs.CV↗

arXiv2025

Zero-Shot Subject-Centric Generation for Creative Application Using Entropy Fusion

Generative models are widely used in visual content creation. However, current text-to-image models often face challenges in practical applications-such as textile pattern design and meme generation-due to the presence of unwanted elements that are difficult to separate with existing methods. Meanwhile, subject-reference generation has emerged as a key research trend, highlighting the need for techniques that can produce clean, high-quality subject images while effectively removing extraneous components. To address this challenge, we introduce a framework for reliable subject-centric image generation. In this work, we propose an entropy-based feature-weighted fusion method to merge the informative cross-attention features obtained from each sampling step of the pretrained text-to-image model FLUX, enabling a precise mask prediction and subject-centric generation. Additionally, we have developed an agent framework based on Large Language Models (LLMs) that translates users' casual inputs into more descriptive prompts, leading to highly detailed image generation. Simultaneously, the agents extract primary elements of prompts to guide the entropy-based feature fusion, ensuring focused primary element generation without extraneous components. Experimental results and user studies demonstrate our methods generates high-quality subject-centric images, outperform existing methods or other possible pipelines, highlighting the effectiveness of our approach.

cs.CV↗

arXiv2025

Analyzing mixed construction and demolition waste in material recovery facilities: evolution, challenges, and applications of computer vision and deep learning

Improving the automatic and timely recognition of construction and demolition waste composition is crucial for enhancing business returns, economic outcomes and sustainability. While deep learning models show promise in recognizing and classifying homogenous materials, the current literature lacks research assessing their performance for mixed, contaminated material in commercial material recycling facility settings. Despite the increasing numbers of deep learning models and datasets generated in this area, the sub-domain of deep learning analysis of construction and demolition waste piles remains underexplored. To address this gap, recent deep learning algorithms and techniques were explored. This review examines the progression in datasets, sensors and the evolution from object detection towards real-time segmentation models. It also synthesizes research from the past five years on deep learning for construction and demolition waste management, highlighting recent advancements while acknowledging limitations that hinder widespread commercial adoption. The analysis underscores the critical requirement for diverse and high-fidelity datasets, advanced sensor technologies, and robust algorithmic frameworks to facilitate the effective integration of deep learning methodologies into construction and demolition waste management systems. This integration is envisioned to contribute significantly towards the advancement of a more sustainable and circular economic model.

cs.CV↗

arXiv2025

Learning Mask Invariant Mutual Information for Masked Image Modeling

Masked autoencoders (MAEs) represent a prominent self-supervised learning paradigm in computer vision. Despite their empirical success, the underlying mechanisms of MAEs remain insufficiently understood. Recent studies have attempted to elucidate the functioning of MAEs through contrastive learning and feature representation analysis, yet these approaches often provide only implicit insights. In this paper, we propose a new perspective for understanding MAEs by leveraging the information bottleneck principle in information theory. Our theoretical analyses reveal that optimizing the latent features to balance relevant and irrelevant information is key to improving MAE performance. Building upon our proofs, we introduce MI-MAE, a novel method that optimizes MAEs through mutual information maximization and minimization. By enhancing latent features to retain maximal relevant information between them and the output, and minimizing irrelevant information between them and the input, our approach achieves better performance. Extensive experiments on standard benchmarks show that MI-MAE significantly outperforms MAE models in tasks such as image classification, object detection, and semantic segmentation. Our findings validate the theoretical framework and highlight the practical advantages of applying the information bottleneck principle to MAEs, offering deeper insights for developing more powerful self-supervised learning models.

cs.CV↗

arXiv2025

Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage

The widespread deployment of deep learning models in privacy-sensitive domains has amplified concerns regarding privacy risks, particularly those stemming from gradient leakage during training. Current privacy assessments primarily rely on post-training attack simulations. However, these methods are inherently reactive, unable to encompass all potential attack scenarios, and often based on idealized adversarial assumptions. These limitations underscore the need for proactive approaches to privacy risk assessment during the training process. To address this gap, we propose the concept of privacy tokens, which are derived directly from private gradients during training. Privacy tokens encapsulate gradient features and, when combined with data features, offer valuable insights into the extent of private information leakage from training data, enabling real-time measurement of privacy risks without relying on adversarial attack simulations. Additionally, we employ Mutual Information (MI) as a robust metric to quantify the relationship between training data and gradients, providing precise and continuous assessments of privacy leakage throughout the training process. Extensive experiments validate our framework, demonstrating the effectiveness of privacy tokens and MI in identifying and quantifying privacy risks. This proactive approach marks a significant advancement in privacy monitoring, promoting the safer deployment of deep learning models in sensitive applications.

cs.LG↗

arXiv2025

Electronic origin of stability of 2D 1H-phase Janus transition metal dichalcogenides and beyond

Janus transition metal dichalcogenides (JTMDs) monolayers have emerged as a new paradigm to broaden the family of two-dimensional (2D) materials. Despite numerous theoretical predictions of JTMDs, their experimental realization remains scarce, most probably due to intrinsic structural fragility. We identify a dependence of the structural stability of 1H-phase JTMDs on the transition metal group, with Group-VIB-based monolayers exhibiting robust stability, as evidenced by the successful synthesized MoSSe and WSSe. The group-dependent stability arises from the competition between metal-ligand ionic bonding and ligand-ligand covalent bonding, as well as the high-energy d-electron orbital splitting. We propose an electron configuration that describes the interactions of electrons near the Fermi level to correlate the stability, and introduce an electron compensation strategy to stabilize certain unstable JTMDs systems. Guided by the electronic origin of stability, we predict a family of stable 2D Janus transition metal halides with intrinsic ferromagnetic valley properties. This work bridges the gap between electronic structure and stability predictions, and extends the design rules for synthesizing 2D Janus materials.

cond-mat.mtrl-sci↗

arXiv2025

Median of Forests for Robust Density Estimation

Robust density estimation refers to the consistent estimation of the density function even when the data is contaminated by outliers. We find that existing forest density estimation at a certain point is inherently resistant to the outliers outside the cells containing the point, which we call \textit{non-local outliers}, but not resistant to the rest \textit{local outliers}. To achieve robustness against all outliers, we propose an ensemble learning algorithm called \textit{medians of forests for robust density estimation} (\textit{MFRDE}), which adopts a pointwise median operation on forest density estimators fitted on subsampled datasets. Compared to existing robust kernel-based methods, MFRDE enables us to choose larger subsampling sizes, sacrificing less accuracy for density estimation while achieving robustness. On the theoretical side, we introduce the local outlier exponent to quantify the number of local outliers. Under this exponent, we show that even if the number of outliers reaches a certain polynomial order in the sample size, MFRDE is able to achieve almost the same convergence rate as the same algorithm on uncontaminated data, whereas robust kernel-based methods fail. On the practical side, real data experiments show that MFRDE outperforms existing robust kernel-based methods. Moreover, we apply MFRDE to anomaly detection to showcase a further application.

stat.ML↗

arXiv2025

Multiscale simulation of rarefied polyatomic gas flow via DIG method

A novel multiscale numerical method is developed to accelerate direct simulation Monte Carlo (DSMC) simulations for polyatomic gases with internal energy. This approach applies the general synthetic iterative scheme to stochastic simulations, removing the inherent restrictions on spatial grid size in DSMC and boosting the evolution of the particle distribution towards a steady state. Firstly, the proposed method intermittently couples the standard DSMC solver with the solution of steady-state macroscopic synthetic equations derived from the kinetic equation for polyatomic gases. These synthetic equations, encompassing accurate constitutive relations and relaxation rates, ensure applicability across the entire flow regime. Secondly, the particle distribution within the DSMC framework is adjusted to align with the solution of the macroscopic equations, significantly accelerating convergence to the steady-state solution. Finally, an adaptive treatment is implemented for the constitutive relations. Higher-order terms extracted from DSMC are applied only in rarefied regions where the local Knudsen number exceeds a predefined threshold. The fast convergence and asymptotic-preserving properties of the proposed method are demonstrated through several numerical simulations, in which it achieves remarkable speedups compared to standard DSMC, particularly in the near-continuum regime.

physics.comp-ph↗

arXiv2025

Stationary solutions with vacuum for a hyperbolic-parabolic chemotaxis model in dimension two

In this research, we study the existence of stationary solutions with vacuum to a hyperbolic-parabolic chemotaxis model with nonlinear pressure in dimension two that describes vasculogenesis. We seek solutions in the radial symmetric class of the whole space, in which the system will be reduced to a system of ODE's on $(0,\infty)$. The fundamental solutions to the ODE system are the Bessel functions of different types. We find two nontrivial solutions. One is formed by half bump (positive density region) starting at $r=0$ and a region of vacuum on the right. Another one is a full nonsymmetric bump away from $r=0$. These solutions bear certain resemblance to in vitro vascular network and the numerically produced structure by Gamba et al arXiv:cond-mat/0303468v1. We also show the nonexistence of full bump starting at $r=0$ and nonexistence of full symmetric bump away from $r=0$.

math.AP↗

arXiv2025

MiniMax-01: Scaling Foundation Models with Lightning Attention

We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.

cs.CL↗

arXiv2024

Adaptive Training Meets Progressive Scaling: Elevating Efficiency in Diffusion Models

Diffusion models have demonstrated remarkable efficacy in various generative tasks with the predictive prowess of denoising model. Currently, diffusion models employ a uniform denoising model across all timesteps. However, the inherent variations in data distributions at different timesteps lead to conflicts during training, constraining the potential of diffusion models. To address this challenge, we propose a novel two-stage divide-and-conquer training strategy termed TDC Training. It groups timesteps based on task similarity and difficulty, assigning highly customized denoising models to each group, thereby enhancing the performance of diffusion models. While two-stage training avoids the need to train each model separately, the total training cost is even lower than training a single unified denoising model. Additionally, we introduce Proxy-based Pruning to further customize the denoising models. This method transforms the pruning problem of diffusion models into a multi-round decision-making problem, enabling precise pruning of diffusion models. Our experiments validate the effectiveness of TDC Training, demonstrating improvements in FID of 1.5 on ImageNet64 compared to original IDDPM, while saving about 20\% of computational resources.

cs.LG↗

arXiv2024

Cross-Self KV Cache Pruning for Efficient Vision-Language Inference

KV cache pruning has emerged as a promising technique for reducing memory and computation costs in long-context auto-regressive generation. Existing methods for vision-language models (VLMs) typically rely on self-attention scores from large language models (LLMs) to identify and prune irrelevant tokens. However, these approaches overlook the inherent distributional discrepancies between modalities, often leading to inaccurate token importance estimation and the over-pruning of critical visual tokens. To address this, we propose decomposing attention scores into intra-modality attention (within the same modality) and inter-modality attention (across modalities), enabling more precise KV cache pruning by independently managing these distinct attention types. Additionally, we introduce an n-softmax function to counteract distribution shifts caused by pruning, preserving the original smoothness of attention scores and ensuring stable performance. Our final training-free method, \textbf{C}ross-\textbf{S}elf \textbf{P}runing (CSP), achieves competitive performance compared to models with full KV caches while significantly outperforming previous pruning methods. Extensive evaluations on MileBench, a benchmark encompassing 29 multimodal datasets, demonstrate CSP's effectiveness, achieving up to a 41\% performance improvement on challenging tasks like conversational embodied dialogue while reducing the KV cache budget by 13.6\%. The code is available at https://github.com/TerryPei/CSP

cs.CV↗

arXiv2024

Intermediate Outputs Are More Sensitive Than You Think

The increasing reliance on deep computer vision models that process sensitive data has raised significant privacy concerns, particularly regarding the exposure of intermediate results in hidden layers. While traditional privacy risk assessment techniques focus on protecting overall model outputs, they often overlook vulnerabilities within these intermediate representations. Current privacy risk assessment techniques typically rely on specific attack simulations to assess risk, which can be computationally expensive and incomplete. This paper introduces a novel approach to measuring privacy risks in deep computer vision models based on the Degrees of Freedom (DoF) and sensitivity of intermediate outputs, without requiring adversarial attack simulations. We propose a framework that leverages DoF to evaluate the amount of information retained in each layer and combines this with the rank of the Jacobian matrix to assess sensitivity to input variations. This dual analysis enables systematic measurement of privacy risks at various model layers. Our experimental validation on real-world datasets demonstrates the effectiveness of this approach in providing deeper insights into privacy risks associated with intermediate representations.

cs.CV↗

arXiv2024

Learning Humanoid Locomotion with Perceptive Internal Model

In contrast to quadruped robots that can navigate diverse terrains using a "blind" policy, humanoid robots require accurate perception for stable locomotion due to their high degrees of freedom and inherently unstable morphology. However, incorporating perceptual signals often introduces additional disturbances to the system, potentially reducing its robustness, generalizability, and efficiency. This paper presents the Perceptive Internal Model (PIM), which relies on onboard, continuously updated elevation maps centered around the robot to perceive its surroundings. We train the policy using ground-truth obstacle heights surrounding the robot in simulation, optimizing it based on the Hybrid Internal Model (HIM), and perform inference with heights sampled from the constructed elevation map. Unlike previous methods that directly encode depth maps or raw point clouds, our approach allows the robot to perceive the terrain beneath its feet clearly and is less affected by camera movement or noise. Furthermore, since depth map rendering is not required in simulation, our method introduces minimal additional computational costs and can train the policy in 3 hours on an RTX 4090 GPU. We verify the effectiveness of our method across various humanoid robots, various indoor and outdoor terrains, stairs, and various sensor configurations. Our method can enable a humanoid robot to continuously climb stairs and has the potential to serve as a foundational algorithm for the development of future humanoid control methods.

cs.RO↗

arXiv2024

Gradient-Guided Conditional Diffusion Models for Private Image Reconstruction: Analyzing Adversarial Impacts of Differential Privacy and Denoising

We investigate the construction of gradient-guided conditional diffusion models for reconstructing private images, focusing on the adversarial interplay between differential privacy noise and the denoising capabilities of diffusion models. While current gradient-based reconstruction methods struggle with high-resolution images due to computational complexity and prior knowledge requirements, we propose two novel methods that require minimal modifications to the diffusion model's generation process and eliminate the need for prior knowledge. Our approach leverages the strong image generation capabilities of diffusion models to reconstruct private images starting from randomly generated noise, even when a small amount of differentially private noise has been added to the gradients. We also conduct a comprehensive theoretical analysis of the impact of differential privacy noise on the quality of reconstructed images, revealing the relationship among noise magnitude, the architecture of attacked models, and the attacker's reconstruction capability. Additionally, extensive experiments validate the effectiveness of our proposed methods and the accuracy of our theoretical findings, suggesting new directions for privacy risk auditing using conditional diffusion models.

cs.CV↗

arXiv2024

Enhancing DP-SGD through Non-monotonous Adaptive Scaling Gradient Weight

In the domain of deep learning, the challenge of protecting sensitive data while maintaining model utility is significant. Traditional Differential Privacy (DP) techniques such as Differentially Private Stochastic Gradient Descent (DP-SGD) typically employ strategies like direct or per-sample adaptive gradient clipping. These methods, however, compromise model accuracy due to their critical influence on gradient handling, particularly neglecting the significant contribution of small gradients during later training stages. In this paper, we introduce an enhanced version of DP-SGD, named Differentially Private Per-sample Adaptive Scaling Clipping (DP-PSASC). This approach replaces traditional clipping with non-monotonous adaptive gradient scaling, which alleviates the need for intensive threshold setting and rectifies the disproportionate weighting of smaller gradients. Our contribution is twofold. First, we develop a novel gradient scaling technique that effectively assigns proper weights to gradients, particularly small ones, thus improving learning under differential privacy. Second, we integrate a momentum-based method into DP-PSASC to reduce bias from stochastic sampling, enhancing convergence rates. Our theoretical and empirical analyses confirm that DP-PSASC preserves privacy and delivers superior performance across diverse datasets, setting new standards for privacy-sensitive applications.

cs.LG↗

arXiv2024

Decentralized Social Networks and the Future of Free Speech Online

Decentralized social networks like Mastodon and BlueSky are trending topics that have drawn much attention and discussion in recent years. By devolving powers from the central node to the end users, decentralized social networks aim to cure existing pathologies on the centralized platforms and have been viewed by many as the future of the Internet. This article critically and systematically assesses the decentralization project's prospect for communications online. It uses normative theories of free speech to examine whether and how the decentralization design could facilitate users' freedom of expression online. The analysis shows that both promises and pitfalls exist, highlighting the importance of value-based design in this area. Two most salient issues for the design of the decentralized networks are: how to balance the decentralization ideal with constant needs of centralization on the network, and how to empower users to make them truly capable of exercising their control. The article then uses some design examples, such as the shared blocklist and the opt-in search function, to illustrate the value considerations underlying the design choices. Some tentative proposals for law and policy interventions are offered to better facilitate the design of the new network. Rather than providing clear answers, the article seeks to map the value implications of the design choices, highlight the stakes, and point directions for future research.

cs.CY↗