arXiv subjects

Wei Feng

Publications and source records attributed to Wei Feng.

At least 109 records · Page 6Linked to original sources

arXiv2025

Decision PCR: Decision version of the Point Cloud Registration task

Low-overlap point cloud registration (PCR) remains a significant challenge in 3D vision. Traditional evaluation metrics, such as Maximum Inlier Count, become ineffective under extremely low inlier ratios. In this paper, we revisit the registration result evaluation problem and identify the Decision version of the PCR task as the fundamental problem. To address this Decision PCR task, we propose a data-driven approach. First, we construct a corresponding dataset based on the 3DMatch dataset. Then, a deep learning-based classifier is trained to reliably assess registration quality, overcoming the limitations of traditional metrics. To our knowledge, this is the first comprehensive study to address this task through a deep learning framework. We incorporate this classifier into standard PCR pipelines. When integrated with our approach, existing state-of-the-art PCR methods exhibit significantly enhanced registration performance. For example, combining our framework with GeoTransformer achieves a new SOTA registration recall of 86.97\% on the challenging 3DLoMatch benchmark. Our method also demonstrates strong generalization capabilities on the unseen outdoor ETH dataset.

cs.CV↗

arXiv2025

Probing nontrivial fusion of Majorana zero modes via near-adiabatic coupling

We propose and simulate a near-adiabatically coupling probing scheme for nontrivial fusion of a pair of Majorara zero modes (MZMs). The scheme can avoid the complexity of oscillating charge occupation in the probing quantum dot, making thus practical measurements more feasible. We also show how to extract the information of nonadiabatic transition and fermion parity violation caused during moving the MZMs together to fuse, from the initial states prepared with definite fermion parity. All the simulations, including the effective coupling between the fusing MZMs, and their coupling to the probing quantum dot, are based on the lattice model of a Rashba quantum wire in proximity contact with an s-wave superconductor, under the modulation of mini-gate voltage control.

cond-mat.mes-hall↗

arXiv2025

Third-order exceptional surface in a pseudo-Hermitian superconducting circuit

Compared with an isolated exceptional point, exceptional surfaces in non-Hermitian systems are more robust against environment noises, fabrication errors, and experimental uncertainties. Thanks to this, exceptional surfaces can be applied to enhance the sensitivity of sensors and develop new quantum techniques. Over the past few years, several works have been devoted to studying high-order exceptional surfaces. However, they are restricted to non-Hermitian systems without pseudo-Hermiticity. To date, research on high-order exceptional surfaces in pseudo-Hermitian systems still remains an untouched area. In this work, we propose a pseudo-Hermitian superconducting circuit, which consists of three circularly-coupled superconducting cavities with the balanced gain and loss. We then study the third-order exceptional surface in the proposed circuit. By investigating the eigenvalues, we find that in the parameter space, all third-order exceptional points of the circuit form a third-order exceptional line in the parity-time-symmetric case. When the parity-time-symmetric condition is extended to pseudo-Hermitian conditions, we find more third-order exceptional points, which constitute a third-order exceptional surface in the parameter space. The proposed scheme is universal and can be applied to explore third-order exceptional surfaces in other physical systems, such as optomechanical systems, cavity-magnon systems, and photonic micro-ring systems. This work is of fundamental interest in quantum mechanics and opens a way for studying high-order exceptional surfaces in pseudo-Hermitian systems.

quant-ph↗

arXiv2025

A novel fast short-time root music method for vibration monitoring of high-speed spindles

Ultra-high-speed spindle bearings challenge traditional vibration monitoring due to broadband noise, non-stationarity, and limited time-frequency resolution. We present a fast Short-Time Root-MUSIC (fSTrM) algorithm that exploits FFT-accelerated Lanczos bidiagonalization to reduce computational complexity from $\mathcal{O}(N^3)$ to $SN\log_2N+S^2(N+S)+M^2(N+M)$ while preserving parametric super-resolution. The method constructs Hankel matrices from 16 ms signal frames and extracts fault frequencies through polynomial rooting on the unit circle. Experimental validation on the Politecnico di Torino bearing dataset demonstrates breakthrough micro-defect detection capabilities. The algorithm reliably identifies 150 $μ$m defects -- previously undetectable by conventional methods -- providing 72+ hours additional warning time. Compared to STFT and wavelet methods, fSTrM achieves 1.2 Hz frequency resolution (vs. 12.5 Hz), 93\% detection rate at $-$5 dB SNR, and quantifies defect severity through harmonic content analysis. Critically, the algorithm processes each frame in 2.4 ms on embedded ARM Cortex-M7 hardware, enabling real-time deployment. This advancement transforms bearing monitoring from failure prevention to continuous degradation assessment, establishing a new paradigm for predictive maintenance in aerospace and precision machining.

cs.IR↗

arXiv2025

TorchTitan: One-stop PyTorch native solution for production ready LLM pre-training

The development of large language models (LLMs) has been instrumental in advancing state-of-the-art natural language processing applications. Training LLMs with billions of parameters and trillions of tokens require sophisticated distributed systems that enable composing and comparing several state-of-the-art techniques in order to efficiently scale across thousands of accelerators. However, existing solutions are complex, scattered across multiple libraries/repositories, lack interoperability, and are cumbersome to maintain. Thus, curating and empirically comparing training recipes require non-trivial engineering effort. This paper introduces TorchTitan, an open-source, PyTorch-native distributed training system that unifies state-of-the-art techniques, streamlining integration and reducing overhead. TorchTitan enables 3D parallelism in a modular manner with elastic scaling, providing comprehensive logging, checkpointing, and debugging tools for production-ready training. It also incorporates hardware-software co-designed solutions, leveraging features like Float8 training and SymmetricMemory. As a flexible test bed, TorchTitan facilitates custom recipe curation and comparison, allowing us to develop optimized training recipes for Llama 3.1 and provide guidance on selecting techniques for maximum efficiency based on our experiences. We thoroughly assess TorchTitan on the Llama 3.1 family of LLMs, spanning 8 billion to 405 billion parameters, and showcase its exceptional performance, modular composability, and elastic scalability. By stacking training optimizations, we demonstrate accelerations of 65.08% with 1D parallelism at the 128-GPU scale (Llama 3.1 8B), an additional 12.59% with 2D parallelism at the 256-GPU scale (Llama 3.1 70B), and an additional 30% with 3D parallelism at the 512-GPU scale (Llama 3.1 405B) on NVIDIA H100 GPUs over optimized baselines.

cs.CL↗

arXiv2025

Enhancing Interpretable Image Classification Through LLM Agents and Conditional Concept Bottleneck Models

Concept Bottleneck Models (CBMs) decompose image classification into a process governed by interpretable, human-readable concepts. Recent advances in CBMs have used Large Language Models (LLMs) to generate candidate concepts. However, a critical question remains: What is the optimal number of concepts to use? Current concept banks suffer from redundancy or insufficient coverage. To address this issue, we introduce a dynamic, agent-based approach that adjusts the concept bank in response to environmental feedback, optimizing the number of concepts for sufficiency yet concise coverage. Moreover, we propose Conditional Concept Bottleneck Models (CoCoBMs) to overcome the limitations in traditional CBMs' concept scoring mechanisms. It enhances the accuracy of assessing each concept's contribution to classification tasks and feature an editable matrix that allows LLMs to correct concept scores that conflict with their internal knowledge. Our evaluations across 6 datasets show that our method not only improves classification accuracy by 6% but also enhances interpretability assessments by 30%.

cs.CL↗

arXiv2025

NUC-Net: Non-uniform Cylindrical Partition Network for Efficient LiDAR Semantic Segmentation

LiDAR semantic segmentation plays a vital role in autonomous driving. Existing voxel-based methods for LiDAR semantic segmentation apply uniform partition to the 3D LiDAR point cloud to form a structured representation based on cartesian/cylindrical coordinates. Although these methods show impressive performance, the drawback of existing voxel-based methods remains in two aspects: (1) it requires a large enough input voxel resolution, which brings a large amount of computation cost and memory consumption. (2) it does not well handle the unbalanced point distribution of LiDAR point cloud. In this paper, we propose a non-uniform cylindrical partition network named NUC-Net to tackle the above challenges. Specifically, we propose the Arithmetic Progression of Interval (API) method to non-uniformly partition the radial axis and generate the voxel representation which is representative and efficient. Moreover, we propose a non-uniform multi-scale aggregation method to improve contextual information. Our method achieves state-of-the-art performance on SemanticKITTI and nuScenes datasets with much faster speed and much less training time. And our method can be a general component for LiDAR semantic segmentation, which significantly improves both the accuracy and efficiency of the uniform counterpart by $4 \times$ training faster and $2 \times$ GPU memory reduction and $3 \times$ inference speedup. We further provide theoretical analysis towards understanding why NUC is effective and how point distribution affects performance. Code is available at \href{https://github.com/alanWXZ/NUC-Net}{https://github.com/alanWXZ/NUC-Net}.

cs.CV↗

arXiv2025

Quantum dissipative dynamics of driven Duffing oscillator near attractors

We investigate the quantum dissipative dynamics near the stable states (attractors) of a driven Duffing oscillator. A refined perturbation theory that can treat two perturbative parameters with different orders is developed to calculate the quantum properties of the Duffing oscillator near the attractors. We obtain the perturbative analytical results, that go beyond the standard linearization approach, for the renormalized level spacings, the orbital displacements, and the effective temperature near the classical attractor. Furthermore, we demonstrate that strong damping induces additional slight renormalization of level spacings, and the Bose distribution together with dephasing. Our work provides new insights into the quantum dynamics of the driven Duffing oscillator and offers a theoretical framework that can be applied to related quantum systems near their stable states.

quant-ph↗

arXiv2025

LLM-Enabled Style and Content Regularization for Personalized Text-to-Image Generation

The personalized text-to-image generation has rapidly advanced with the emergence of Stable Diffusion. Existing methods, which typically fine-tune models using embedded identifiers, often struggle with insufficient stylization and inaccurate image content due to reduced textual controllability. In this paper, we propose style refinement and content preservation strategies. The style refinement strategy leverages the semantic information of visual reasoning prompts and reference images to optimize style embeddings, allowing a more precise and consistent representation of style information. The content preservation strategy addresses the content bias problem by preserving the model's generalization capabilities, ensuring enhanced textual controllability without compromising stylization. Experimental results verify that our approach achieves superior performance in generating consistent and personalized text-to-image outputs.

cs.CV↗

arXiv2025

Beyond Background Shift: Rethinking Instance Replay in Continual Semantic Segmentation

In this work, we focus on continual semantic segmentation (CSS), where segmentation networks are required to continuously learn new classes without erasing knowledge of previously learned ones. Although storing images of old classes and directly incorporating them into the training of new models has proven effective in mitigating catastrophic forgetting in classification tasks, this strategy presents notable limitations in CSS. Specifically, the stored and new images with partial category annotations leads to confusion between unannotated categories and the background, complicating model fitting. To tackle this issue, this paper proposes a novel Enhanced Instance Replay (EIR) method, which not only preserves knowledge of old classes while simultaneously eliminating background confusion by instance storage of old classes, but also mitigates background shifts in the new images by integrating stored instances with new images. By effectively resolving background shifts in both stored and new images, EIR alleviates catastrophic forgetting in the CSS task, thereby enhancing the model's capacity for CSS. Experimental results validate the efficacy of our approach, which significantly outperforms state-of-the-art CSS methods.

cs.CV↗

arXiv2025

Casual Inference via Style Bias Deconfounding for Domain Generalization

Deep neural networks (DNNs) often struggle with out-of-distribution data, limiting their reliability in diverse realworld applications. To address this issue, domain generalization methods have been developed to learn domain-invariant features from single or multiple training domains, enabling generalization to unseen testing domains. However, existing approaches usually overlook the impact of style frequency within the training set. This oversight predisposes models to capture spurious visual correlations caused by style confounding factors, rather than learning truly causal representations, thereby undermining inference reliability. In this work, we introduce Style Deconfounding Causal Learning (SDCL), a novel causal inference-based framework designed to explicitly address style as a confounding factor. Our approaches begins with constructing a structural causal model (SCM) tailored to the domain generalization problem and applies a backdoor adjustment strategy to account for style influence. Building on this foundation, we design a style-guided expert module (SGEM) to adaptively clusters style distributions during training, capturing the global confounding style. Additionally, a back-door causal learning module (BDCL) performs causal interventions during feature extraction, ensuring fair integration of global confounding styles into sample predictions, effectively reducing style bias. The SDCL framework is highly versatile and can be seamlessly integrated with state-of-the-art data augmentation techniques. Extensive experiments across diverse natural and medical image recognition tasks validate its efficacy, demonstrating superior performance in both multi-domain and the more challenging single-domain generalization scenarios.

cs.CV↗

arXiv2025

CTR-Driven Advertising Image Generation with Multimodal Large Language Models

In web data, advertising images are crucial for capturing user attention and improving advertising effectiveness. Most existing methods generate background for products primarily focus on the aesthetic quality, which may fail to achieve satisfactory online performance. To address this limitation, we explore the use of Multimodal Large Language Models (MLLMs) for generating advertising images by optimizing for Click-Through Rate (CTR) as the primary objective. Firstly, we build targeted pre-training tasks, and leverage a large-scale e-commerce multimodal dataset to equip MLLMs with initial capabilities for advertising image generation tasks. To further improve the CTR of generated images, we propose a novel reward model to fine-tune pre-trained MLLMs through Reinforcement Learning (RL), which can jointly utilize multimodal features and accurately reflect user click preferences. Meanwhile, a product-centric preference optimization strategy is developed to ensure that the generated background content aligns with the product characteristics after fine-tuning, enhancing the overall relevance and effectiveness of the advertising images. Extensive experiments have demonstrated that our method achieves state-of-the-art performance in both online and offline metrics. Our code and pre-trained models are publicly available at: https://github.com/Chenguoz/CAIG.

cs.LG↗

arXiv2025

Synthetic-To-Real Video Person Re-ID

Person re-identification (Re-ID) is an important task and has significant applications for public security and information forensics, which has progressed rapidly with the development of deep learning. In this work, we investigate a novel and challenging setting of Re-ID, i.e., cross-domain video-based person Re-ID. Specifically, we utilize synthetic video datasets as the source domain for training and real-world videos for testing, notably reducing the reliance on expensive real data acquisition and annotation. To harness the potential of synthetic data, we first propose a self-supervised domain-invariant feature learning strategy for both static and dynamic (temporal) features. Additionally, to enhance person identification accuracy in the target domain, we propose a mean-teacher scheme incorporating a self-supervised ID consistency loss. Experimental results across five real datasets validate the rationale behind cross-synthetic-real domain adaptation and demonstrate the efficacy of our method. Notably, the discovery that synthetic data outperforms real data in the cross-domain scenario is a surprising outcome. The code and data are publicly available at https://github.com/XiangqunZhang/UDA_Video_ReID.

cs.CV↗

arXiv2025

A Block Term Decomposition Model Based Algorithm for Tensor Completion of Multidimensional Harmonic Signals

We consider tensor data completion of an incomplete observation of multidimensional harmonic (MH) signals. Unlike existing tensor-based techniques for MH retrieval (MHR), which mostly adopt the canonical polyadic decomposition (CPD) to model the simple "one-to-one" correspondence among harmonics across difference modes, we herein use the more flexible block term decomposition (BTD) model that can be used to describe the complex mutual correspondences among several groups of harmonics across different modes. An optimization principle that aims to fit the BTD model in the least squares sense, subject to rank minimization of hankelized MH components, is set up for the tensor completion task, and an algorithm based on alternating direction method of multipliers is proposed, of which the effectiveness and applicability are validated through both numerical simulations and an application in Sub-6GHz channel state information (CSI) completion.

eess.SP↗

arXiv2025

A parametric non-negative coupled canonical polyadic decomposition algorithm for hyperspectral super-resolution

Recently, coupled tensor decomposition has been widely used in data fusion of a hyperspectral image (HSI) and a multispectral image (MSI) for hyperspectral super-resolution (HSR). However, exsiting works often ignore the inherent non-negative (NN) property of the image data, or impose the NN constraint via hard-thresholding which may interfere with the optimization procedure and cause the method to be sub-optimal. As such, we propose a novel NN coupled canonical polyadic decomposition (NN-C-CPD) algorithm, which makes use of the parametric method and nonlinear least squares (NLS) framework to impose the NN constraint into the C-CPD computation. More exactly, the NN constraint is converted into the squared relationship between the NN entries of the factor matrices and a set of latent parameters. Based on the chain rule for deriving the derivatives, the key entities such as gradient and Jacobian with regards to the latent parameters can be derived, thus the NN constraint is naturally integrated without interfering with the optimization procedure. Experimental results are provided to demonstrate the performance of the proposed NN-C-CPD algorithm in HSR applications.

eess.SP↗

arXiv2025

Generate E-commerce Product Background by Integrating Category Commonality and Personalized Style

The state-of-the-art methods for e-commerce product background generation suffer from the inefficiency of designing product-wise prompts when scaling up the production, as well as the ineffectiveness of describing fine-grained styles when customizing personalized backgrounds for some specific brands. To address these obstacles, we integrate the category commonality and personalized style into diffusion models. Concretely, we propose a Category-Wise Generator to enable large-scale background generation with only one model for the first time. A unique identifier in the prompt is assigned to each category, whose attention is located on the background by a mask-guided cross attention layer to learn the category-wise style. Furthermore, for products with specific and fine-grained requirements in layout, elements, etc, a Personality-Wise Generator is devised to learn such personalized style directly from a reference image to resolve textual ambiguities, and is trained in a self-supervised manner for more efficient training data usage. To advance research in this field, the first large-scale e-commerce product background generation dataset BG60k is constructed, which covers more than 60k product images from over 2k categories. Experiments demonstrate that our method could generate high-quality backgrounds for different categories, and maintain the personalized background style of reference images. BG60k will be available at \url{https://github.com/Whileherham/BG60k}.

cs.CV↗

arXiv2025

Investigating Numerical Translation with Large Language Models

The inaccurate translation of numbers can lead to significant security issues, ranging from financial setbacks to medical inaccuracies. While large language models (LLMs) have made significant advancements in machine translation, their capacity for translating numbers has not been thoroughly explored. This study focuses on evaluating the reliability of LLM-based machine translation systems when handling numerical data. In order to systematically test the numerical translation capabilities of currently open source LLMs, we have constructed a numerical translation dataset between Chinese and English based on real business data, encompassing ten types of numerical translation. Experiments on the dataset indicate that errors in numerical translation are a common issue, with most open-source LLMs faltering when faced with our test scenarios. Especially when it comes to numerical types involving large units like ``million", ``billion", and "yi", even the latest llama3.1 8b model can have error rates as high as 20%. Finally, we introduce three potential strategies to mitigate the numerical mistranslations for large units.

cs.CL↗

arXiv2024

CCxTrust: Confidential Computing Platform Based on TEE and TPM Collaborative Trust

Confidential Computing has emerged to address data security challenges in cloud-centric deployments by protecting data in use through hardware-level isolation. However, reliance on a single hardware root of trust (RoT) limits user confidence in cloud platforms, especially for high-performance AI services, where end-to-end protection of sensitive models and data is critical. Furthermore, the lack of interoperability and a unified trust model in multi-cloud environments prevents the establishment of a cross-platform, cross-cloud chain of trust, creating a significant trust gap for users with high privacy requirements. To address the challenges mentioned above, this paper proposes CCxTrust (Confidential Computing with Trust), a confidential computing platform leveraging collaborative roots of trust from TEE and TPM. CCxTrust combines the black-box RoT embedded in the CPU-TEE with the flexible white-box RoT of TPM to establish a collaborative trust framework. The platform implements independent Roots of Trust for Measurement (RTM) for TEE and TPM, and a collaborative Root of Trust for Report (RTR) for composite attestation. The Root of Trust for Storage (RTS) is solely supported by TPM. We also present the design and implementation of a confidential TPM supporting multiple modes for secure use within confidential virtual machines. Additionally, we propose a composite attestation protocol integrating TEE and TPM to enhance security and attestation efficiency, which is proven secure under the PCL protocol security model. We implemented a prototype of CCxTrust on a confidential computing server with AMD SEV-SNP and TPM chips, requiring minimal modifications to the TPM and guest Linux kernel. The composite attestation efficiency improved by 24% without significant overhead, while Confidential TPM performance showed a 16.47% reduction compared to standard TPM.

cs.CR↗