Searcharxiv⌕ Search

arXiv subjects

Ze Yang

Publications and source records attributed to Ze Yang.

At least 37 records · Page 2Linked to original sources
arXiv2025

Scalability Optimization in Cloud-Based AI Inference Services: Strategies for Real-Time Load Balancing and Automated Scaling

The rapid expansion of AI inference services in the cloud necessitates a robust scalability solution to manage dynamic workloads and maintain high performance. This study proposes a comprehensive scalability optimization framework for cloud AI inference services, focusing on real-time load balancing and autoscaling strategies. The proposed model is a hybrid approach that combines reinforcement learning for adaptive load distribution and deep neural networks for accurate demand forecasting. This multi-layered approach enables the system to anticipate workload fluctuations and proactively adjust resources, ensuring maximum resource utilisation and minimising latency. Furthermore, the incorporation of a decentralised decision-making process within the model serves to enhance fault tolerance and reduce response time in scaling operations. Experimental results demonstrate that the proposed model enhances load balancing efficiency by 35\ and reduces response delay by 28\, thereby exhibiting a substantial optimization effect in comparison with conventional scalability solutions.

cs.DC↗
arXiv2025

NeRF-Based defect detection

The rapid growth of industrial automation has highlighted the need for precise and efficient defect detection in large-scale machinery. Traditional inspection techniques, involving manual procedures such as scaling tall structures for visual evaluation, are labor-intensive, subjective, and often hazardous. To overcome these challenges, this paper introduces an automated defect detection framework built on Neural Radiance Fields (NeRF) and the concept of digital twins. The system utilizes UAVs to capture images and reconstruct 3D models of machinery, producing both a standard reference model and a current-state model for comparison. Alignment of the models is achieved through the Iterative Closest Point (ICP) algorithm, enabling precise point cloud analysis to detect deviations that signify potential defects. By eliminating manual inspection, this method improves accuracy, enhances operational safety, and offers a scalable solution for defect detection. The proposed approach demonstrates great promise for reliable and efficient industrial applications.

cs.CV↗
arXiv2025

Scam Detection for Ethereum Smart Contracts: Leveraging Graph Representation Learning for Secure Blockchain

As more and more attacks have been detected on Ethereum smart contracts, it has seriously affected finance and credibility. Current anti-fraud detection techniques, including code parsing or manual feature extraction, still have some shortcomings, although some generalization or adaptability can be obtained. In the face of this situation, this paper proposes to use graphical representation learning technology to find transaction patterns and distinguish malicious transaction contracts, that is, to represent Ethereum transaction data as graphs, and then use advanced ML technology to obtain reliable and accurate results. Taking into account the sample imbalance, we treated with SMOTE-ENN and tested several models, in which MLP performed better than GCN, but the exact effect depends on its field trials. Our research opens up more possibilities for trust and security in the Ethereum ecosystem.

cs.LG↗
arXiv2025

Research on Large Language Model Cross-Cloud Privacy Protection and Collaborative Training based on Federated Learning

The fast development of large language models (LLMs) and popularization of cloud computing have led to increasing concerns on privacy safeguarding and data security of cross-cloud model deployment and training as the key challenges. We present a new framework for addressing these issues along with enabling privacy preserving collaboration on training between distributed clouds based on federated learning. Our mechanism encompasses cutting-edge cryptographic primitives, dynamic model aggregation techniques, and cross-cloud data harmonization solutions to enhance security, efficiency, and scalability to the traditional federated learning paradigm. Furthermore, we proposed a hybrid aggregation scheme to mitigate the threat of Data Leakage and to optimize the aggregation of model updates, thus achieving substantial enhancement on the model effectiveness and stability. Experimental results demonstrate that the training efficiency, privacy protection, and model accuracy of the proposed model compare favorably to those of the traditional federated learning method.

cs.CR↗
arXiv2025

Adaptive Fault Tolerance Mechanisms of Large Language Models in Cloud Computing Environments

With the rapid evolution of Large Language Models (LLMs) and their large-scale experimentation in cloud-computing spaces, the challenge of guaranteeing their security and efficiency in a failure scenario has become a main issue. To ensure the reliability and availability of large-scale language models in cloud computing scenarios, such as frequent resource failures, network problems, and computational overheads, this study proposes a novel adaptive fault tolerance mechanism. It builds upon known fault-tolerant mechanisms, such as checkpointing, redundancy, and state transposition, introducing dynamic resource allocation and prediction of failure based on real-time performance metrics. The hybrid model integrates data driven deep learning-based anomaly detection technique underlining the contribution of cloud orchestration middleware for predictive prevention of system failures. Additionally, the model integrates adaptive checkpointing and recovery strategies that dynamically adapt according to load and system state to minimize the influence on the performance of the model and minimize downtime. The experimental results demonstrate that the designed model considerably enhances the fault tolerance in large-scale cloud surroundings, and decreases the system downtime by $\mathbf{30\%}$, and has a better modeling availability than the classical fault tolerance mechanism.

cs.DC↗
arXiv2025

Advances in APPFL: A Comprehensive and Extensible Federated Learning Framework

Federated learning (FL) is a distributed machine learning paradigm enabling collaborative model training while preserving data privacy. In today's landscape, where most data is proprietary, confidential, and distributed, FL has become a promising approach to leverage such data effectively, particularly in sensitive domains such as medicine and the electric grid. Heterogeneity and security are the key challenges in FL, however, most existing FL frameworks either fail to address these challenges adequately or lack the flexibility to incorporate new solutions. To this end, we present the recent advances in developing APPFL, an extensible framework and benchmarking suite for federated learning, which offers comprehensive solutions for heterogeneity and security concerns, as well as user-friendly interfaces for integrating new algorithms or adapting to new applications. We demonstrate the capabilities of APPFL through extensive experiments evaluating various aspects of FL, including communication efficiency, privacy preservation, computational performance, and resource utilization. We further highlight the extensibility of APPFL through case studies in vertical, hierarchical, and decentralized FL. APPFL is fully open-sourced on GitHub at https://github.com/APPFL/APPFL.

cs.LG↗
arXiv2025

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model

Large Language Models (LLMs) with API-calling capabilities enabled building effective Language Agents (LA), while also revolutionizing the conventional task-oriented dialogue (TOD) paradigm. However, current approaches face a critical dilemma: TOD systems are often trained on a limited set of target APIs, requiring new data to maintain their quality when interfacing with new services, while LAs are not trained to maintain user intent over multi-turn conversations. Because both robust multi-turn management and advanced function calling are crucial for effective conversational agents, we evaluate these skills on three popular benchmarks: MultiWOZ 2.4 (TOD), BFCL V3 (LA), and API-Bank (LA), and our analyses reveal that specialized approaches excel in one domain but underperform in the other. To bridge this chasm, we introduce CoALM (Conversational Agentic Language Model), a unified approach that integrates both conversational and agentic capabilities. We created CoALM-IT, a carefully constructed multi-task dataset that interleave multi-turn ReAct reasoning with complex API usage. Using CoALM-IT, we train three models CoALM 8B, CoALM 70B, and CoALM 405B, which outperform top domain-specific models, including GPT-4o, across all three benchmarks. This demonstrates the feasibility of a single model approach for both TOD and LA, setting a new standard for conversational agents.

cs.AI↗
arXiv2025

HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Models

Large Language Models (LLMs) have revolutionized natural language processing by understanding and generating human-like text. However, the increasing demand for more sophisticated LLMs presents significant computational challenges due to their scale and complexity. This paper introduces Hardware Accelerated Decoding (HADES), a novel approach to enhance the performance and energy efficiency of LLMs. We address the design of an LLM accelerator with hardware-level speculative decoding support, a concept not previously explored in existing literature. Our work demonstrates how speculative decoding can significantly improve the efficiency of LLM operations, paving the way for more advanced and practical applications of these models.

cs.CL↗
arXiv2024

Towards Lifelong Few-Shot Customization of Text-to-Image Diffusion

Lifelong few-shot customization for text-to-image diffusion aims to continually generalize existing models for new tasks with minimal data while preserving old knowledge. Current customization diffusion models excel in few-shot tasks but struggle with catastrophic forgetting problems in lifelong generations. In this study, we identify and categorize the catastrophic forgetting problems into two folds: relevant concepts forgetting and previous concepts forgetting. To address these challenges, we first devise a data-free knowledge distillation strategy to tackle relevant concepts forgetting. Unlike existing methods that rely on additional real data or offline replay of original concept data, our approach enables on-the-fly knowledge distillation to retain the previous concepts while learning new ones, without accessing any previous data. Second, we develop an In-Context Generation (ICGen) paradigm that allows the diffusion model to be conditioned upon the input vision context, which facilitates the few-shot generation and mitigates the issue of previous concepts forgetting. Extensive experiments show that the proposed Lifelong Few-Shot Diffusion (LFS-Diffusion) method can produce high-quality and accurate images while maintaining previously learned knowledge.

cs.CV↗
arXiv2024

Open-Structure: Structural Benchmark Dataset for SLAM Algorithms

This paper presents Open-Structure, a novel benchmark dataset for evaluating visual odometry and SLAM methods. Compared to existing public datasets that primarily offer raw images, Open-Structure provides direct access to point and line measurements, correspondences, structural associations, and co-visibility factor graphs, which can be fed to various stages of SLAM pipelines to mitigate the impact of data preprocessing modules in ablation experiments. The dataset comprises two distinct types of sequences from the perspective of scenarios. The first type maintains reasonable observation and occlusion relationships, as these critical elements are extracted from public image-based sequences using our dataset generator. In contrast, the second type consists of carefully designed simulation sequences that enhance dataset diversity by introducing a wide range of trajectories and observations. Furthermore, a baseline is proposed using our dataset to evaluate widely used modules, including camera pose tracking, parametrization, and factor graph optimization, within SLAM systems. By evaluating these state-of-the-art algorithms across different scenarios, we discern each module's strengths and weaknesses in the context of camera tracking and optimization processes. The Open-Structure dataset and baseline system are openly accessible on website: \url{https://open-structure.github.io}, encouraging further research and development in the field of SLAM.

cs.RO↗
arXiv2024

Loop unitary and phase band topological invariant in generic multi-band Chern insulators

Quench dynamics of topological phases have been studied in the past few years and dynamical topological invariants are formulated in different ways. Yet most of these invariants are limited to minimal systems in which Hamiltonians are expanded by Gamma matrices. Here we generalize the dynamical 3-winding-number in two-band systems into the one in generic multi-band Chern insulators and prove that its value is equal to the difference of Chern numbers between post-quench and pre-quench Hamiltonians. Moreover we obtain an expression of this dynamical 3-winding-number represented by gapless fermions in phase bands depending only on the phase and its projectors, so it is generic for the quench of all multi-band Chern insulators. Besides, we obtain a multifold fermion in the phase band in (k, t) space by quenching a three-band model, which cannot happen for two band models.

cond-mat.mes-hall↗
arXiv2024

G3R: Gradient Guided Generalizable Reconstruction

Large scale 3D scene reconstruction is important for applications such as virtual reality and simulation. Existing neural rendering approaches (e.g., NeRF, 3DGS) have achieved realistic reconstructions on large scenes, but optimize per scene, which is expensive and slow, and exhibit noticeable artifacts under large view changes due to overfitting. Generalizable approaches or large reconstruction models are fast, but primarily work for small scenes/objects and often produce lower quality rendering results. In this work, we introduce G3R, a generalizable reconstruction approach that can efficiently predict high-quality 3D scene representations for large scenes. We propose to learn a reconstruction network that takes the gradient feedback signals from differentiable rendering to iteratively update a 3D scene representation, combining the benefits of high photorealism from per-scene optimization with data-driven priors from fast feed-forward prediction methods. Experiments on urban-driving and drone datasets show that G3R generalizes across diverse large scenes and accelerates the reconstruction process by at least 10x while achieving comparable or better realism compared to 3DGS, and also being more robust to large view changes.

cs.CV↗
arXiv2024

UniCal: Unified Neural Sensor Calibration

Self-driving vehicles (SDVs) require accurate calibration of LiDARs and cameras to fuse sensor data accurately for autonomy. Traditional calibration methods typically leverage fiducials captured in a controlled and structured scene and compute correspondences to optimize over. These approaches are costly and require substantial infrastructure and operations, making it challenging to scale for vehicle fleets. In this work, we propose UniCal, a unified framework for effortlessly calibrating SDVs equipped with multiple LiDARs and cameras. Our approach is built upon a differentiable scene representation capable of rendering multi-view geometrically and photometrically consistent sensor observations. We jointly learn the sensor calibration and the underlying scene representation through differentiable volume rendering, utilizing outdoor sensor data without the need for specific calibration fiducials. This "drive-and-calibrate" approach significantly reduces costs and operational overhead compared to existing calibration systems, enabling efficient calibration for large SDV fleets at scale. To ensure geometric consistency across observations from different sensors, we introduce a novel surface alignment loss that combines feature-based registration with neural rendering. Comprehensive evaluations on multiple datasets demonstrate that UniCal outperforms or matches the accuracy of existing calibration approaches while being more efficient, demonstrating the value of UniCal for scalable calibration.

cs.CV↗
arXiv2024

3D-VirtFusion: Synthetic 3D Data Augmentation through Generative Diffusion Models and Controllable Editing

Data augmentation plays a crucial role in deep learning, enhancing the generalization and robustness of learning-based models. Standard approaches involve simple transformations like rotations and flips for generating extra data. However, these augmentations are limited by their initial dataset, lacking high-level diversity. Recently, large models such as language models and diffusion models have shown exceptional capabilities in perception and content generation. In this work, we propose a new paradigm to automatically generate 3D labeled training data by harnessing the power of pretrained large foundation models. For each target semantic class, we first generate 2D images of a single object in various structure and appearance via diffusion models and chatGPT generated text prompts. Beyond texture augmentation, we propose a method to automatically alter the shape of objects within 2D images. Subsequently, we transform these augmented images into 3D objects and construct virtual scenes by random composition. This method can automatically produce a substantial amount of 3D scene data without the need of real data, providing significant benefits in addressing few-shot learning challenges and mitigating long-tailed class imbalances. By providing a flexible augmentation approach, our work contributes to enhancing 3D data diversity and advancing model capabilities in scene understanding tasks.

cs.CV↗
arXiv2024

On the radially deformed Fourier transform

In this paper we consider the kernel of the radially deformed Fourier transform introduced in the context of Clifford analysis in [10]. By adapting the Laplace transform method from [4], we obtain the Laplace domain expressions of the kernel for the cases of $m=2$ and $m > 2$ when $1+c=\frac{1}{n}, n\in \mathbb{N}_0\backslash\{1\}$ with $n$ odd. Moreover, we show that the expressions can be simplified using the Poisson kernel and the generating function of the Gegenbauer polynomials. As a consequence, the inverse formulas are used to get the integral expressions of the kernel in terms of Mittag-Leffler functions.

math.CA↗
arXiv2024

Copilot4D: Learning Unsupervised World Models for Autonomous Driving via Discrete Diffusion

Learning world models can teach an agent how the world works in an unsupervised manner. Even though it can be viewed as a special case of sequence modeling, progress for scaling world models on robotic applications such as autonomous driving has been somewhat less rapid than scaling language models with Generative Pre-trained Transformers (GPT). We identify two reasons as major bottlenecks: dealing with complex and unstructured observation space, and having a scalable generative model. Consequently, we propose Copilot4D, a novel world modeling approach that first tokenizes sensor observations with VQVAE, then predicts the future via discrete diffusion. To efficiently decode and denoise tokens in parallel, we recast Masked Generative Image Transformer as discrete diffusion and enhance it with a few simple changes, resulting in notable improvement. When applied to learning world models on point cloud observations, Copilot4D reduces prior SOTA Chamfer distance by more than 65% for 1s prediction, and more than 50% for 3s prediction, across NuScenes, KITTI Odometry, and Argoverse2 datasets. Our results demonstrate that discrete diffusion on tokenized agent experience can unlock the power of GPT-like unsupervised learning for robotics.

cs.CV↗
arXiv2024

Universal scalefree non-Hermitian skin effect near the Bloch point

The scalefree non-Hermitian skin effect (NHSE) refers to the phenomenon that the localization length of skin modes scales proportionally with system size in non-Hermitian systems. Authors of recent studies have demonstrated that the scalefree NHSE can be induced through various mechanisms, including the critical NHSE, local non-Hermiticity, and the boundary impurity effect. Nevertheless, these methods require careful modeling and precise parameter tuning. In contrast, in this paper, we suggest that the scalefree NHSE is a universal phenomenon, observable in extensive systems if these systems can be described by non-Bloch band theory and host Bloch points on the energy spectrum in the thermodynamic limit. Crucially, we discover that the geometry of the generalized Brillouin zone determines the scaling rule of the localization length, which can scale either linearly or quadratically with the system size. In this paper, we enriches the phenomenon of the scalefree NHSE.

quant-ph↗