arXiv subjects

Lei Wang

Publications and source records attributed to Lei Wang.

At least 325 records · Page 18Linked to original sources

arXiv2025

Thermocapillary migration of an odd viscous droplet on a uniformly heated surface: A lattice Boltzmann study

In this study, the thermocapillary actuation behavior of an odd viscous droplet on a uniformly heated surface is numerically investigated using a phase-field-based lattice Boltzmann method. The numerical results reveal that unlike a conventional viscous droplet that remains stationary on a uniformly heated surface, the presence of odd viscosity converts tangential Marangoni stresses into asymmetric normal stresses along the interface, thereby inducing spontaneous droplet motion. Specifically, when the odd viscosity coefficient is positive (negative), the droplet migrates toward the right (left). Additionally, due to the enhanced interfacial temperature gradient, the droplet migration velocity consistently increases with the contact angle. Further, it is observed that the droplet's migration velocity decreases with an increasing viscosity ratio between the surrounding fluid and the droplet. Finally, as the droplet is placed on an inclined surface, its migration direction and velocity are governed by the interaction between gravity and the odd viscosity-induced force, and in certain cases, the droplet can even climb upward against gravity.

physics.flu-dyn↗

arXiv2025

A Review of Longitudinal Radiology Report Generation: Dataset Composition, Methods, and Performance Evaluation

Chest Xray imaging is a widely used diagnostic tool in modern medicine, and its high utilization creates substantial workloads for radiologists. To alleviate this burden, vision language models are increasingly applied to automate Chest Xray radiology report generation (CXRRRG), aiming for clinically accurate descriptions while reducing manual effort. Conventional approaches, however, typically rely on single images, failing to capture the longitudinal context necessary for producing clinically faithful comparison statements. Recently, growing attention has been directed toward incorporating longitudinal data into CXR RRG, enabling models to leverage historical studies in ways that mirror radiologists diagnostic workflows. Nevertheless, existing surveys primarily address single image CXRRRG and offer limited guidance for longitudinal settings, leaving researchers without a systematic framework for model design. To address this gap, this survey provides the first comprehensive review of longitudinal radiology report generation (LRRG). Specifically, we examine dataset construction strategies, report generation architectures alongside longitudinally tailored designs, and evaluation protocols encompassing both longitudinal specific measures and widely used benchmarks. We further summarize LRRG methods performance, alongside analyses of different ablation studies, which collectively highlight the critical role of longitudinal information and architectural design choices in improving model performance. Finally, we summarize five major limitations of current research and outline promising directions for future development, aiming to lay a foundation for advancing this emerging field.

cs.CV↗

arXiv2025

A 95 GeV Higgs boson and spontaneous CP-violation at the finite temperature

The ATLAS and CMS collaborations reported a diphoton excess in the invariant mass distribution around the 95.4 GeV with a local significance of $3.1σ$. Moreover, there is another $2.3σ$ local excess in the $b\bar{b}$ final state at LEP in the same mass region. A plausible solution is that the Higgs sector is extended to include an additional Higgs boson with a mass of $95.4$ GeV. We study a complex singlet scalar extension of the two-Higgs-doublet model in which the 95.4 GeV Higgs is from the mixing of three CP-even Higgs fields. In addition, the extended Higgs potential can achieve spontaneous CP-violation at the finite temperature and restore CP symmetry at the present temperature of the Universe. We find that the model can simultaneously explain the baryon asymmetry of the Universe, the diphoton and $b\bar{b}$ excesses around the 95.4 GeV while satisfying various relevant constraints including the experiments of collider and electric dipole moment.

hep-ph↗

arXiv2025

95 GeV Higgs boson and nano-Hertz gravitational waves from domain walls in the N2HDM

We explore the diphoton and $b\bar{b}$ excesses at 95.4 GeV, as well as nano-Hertz gravitational waves originating from domain walls, within the framework of the next-to-two-Higgs-doublet model (N2HDM), which extends the two-Higgs-doublet model by introducing a real singlet scalar subject to a discrete $Z_2$ symmetry. The $Z_2$ symmetry is spontaneously broken by the non-zero vacuum expectation value of the singlet scalar, $v_s$, which leads to the formation of domain walls. We discuss two different scenarios: in scenario A, the 95.4 GeV Higgs boson predominantly originates from the singlet field, while in scenario B, it arises mainly from the CP-even components of the Higgs doublets. Taking into account relevant theoretical and experimental constraints, we find that scenario A can fully account for both the diphoton and $b\bar{b}$ excesses at 95.4 GeV within the $1σ$ range. Moreover, the peak amplitude of the gravitational wave spectrum at a peak frequency of $10^{-9}$ Hz can reach $2 \times 10^{-12}$ for $v_s = 100$ TeV. Scenario B only marginally accounts for the diphoton and $b\bar{b}$ excesses at the $1σ$ level, but the peak amplitude of the gravitational wave spectrum at the peak frequency of $10^{-9}$ Hz can reach $6\times 10^{-8}$ for $v_s=100$ TeV. The nano-Hertz gravitational wave signals predicted in both scenarios can be tested by the current and future pulsar timing array projects.

hep-ph↗

arXiv2025

LoRA-FAIR: Federated LoRA Fine-Tuning with Aggregation and Initialization Refinement

Foundation models (FMs) achieve strong performance across diverse tasks with task-specific fine-tuning, yet full parameter fine-tuning is often computationally prohibitive for large models. Parameter-efficient fine-tuning (PEFT) methods like Low-Rank Adaptation (LoRA) reduce this cost by introducing low-rank matrices for tuning fewer parameters. While LoRA allows for efficient fine-tuning, it requires significant data for adaptation, making Federated Learning (FL) an appealing solution due to its privacy-preserving collaborative framework. However, combining LoRA with FL introduces two key challenges: the \textbf{Server-Side Aggregation Bias}, where server-side averaging of LoRA matrices diverges from the ideal global update, and the \textbf{Client-Side Initialization Lag}, emphasizing the need for consistent initialization across rounds. Existing approaches address these challenges individually, limiting their effectiveness. We propose LoRA-FAIR, a novel method that tackles both issues by introducing a correction term on the server, enhancing aggregation efficiency and accuracy. LoRA-FAIR maintains computational and communication efficiency, yielding superior performance over state-of-the-art methods. Experimental results on ViT and MLP-Mixer models across large-scale datasets demonstrate that LoRA-FAIR consistently achieves performance improvements in FL settings.

cs.LG↗

arXiv2025

Adaptive LoRA Experts Allocation and Selection for Federated Fine-Tuning

Large Language Models (LLMs) have demonstrated impressive capabilities across various tasks, but fine-tuning them for domain-specific applications often requires substantial domain-specific data that may be distributed across multiple organizations. Federated Learning (FL) offers a privacy-preserving solution, but faces challenges with computational constraints when applied to LLMs. Low-Rank Adaptation (LoRA) has emerged as a parameter-efficient fine-tuning approach, though a single LoRA module often struggles with heterogeneous data across diverse domains. This paper addresses two critical challenges in federated LoRA fine-tuning: 1. determining the optimal number and allocation of LoRA experts across heterogeneous clients, and 2. enabling clients to selectively utilize these experts based on their specific data characteristics. We propose FedLEASE (Federated adaptive LoRA Expert Allocation and SElection), a novel framework that adaptively clusters clients based on representation similarity to allocate and train domain-specific LoRA experts. It also introduces an adaptive top-$M$ Mixture-of-Experts mechanism that allows each client to select the optimal number of utilized experts. Our extensive experiments on diverse benchmark datasets demonstrate that FedLEASE significantly outperforms existing federated fine-tuning approaches in heterogeneous client settings while maintaining communication efficiency.

cs.LG↗

arXiv2025

Hidden phonon-assisted charge density wave transition in BaFe2Al9 revealed by ultrafast optical spectroscopy

The interplay between electronic and lattice degrees of freedom is fundamental to charge density wave (CDW) formation, yet the microscopic origin often remains elusive. Here, we investigate the transient optical response of the intermetallic compound BaFe2Al9 using polarization-resolved ultrafast optical spectroscopy. We identify a discontinuous sign reversal in the transient reflectivity at Tc ~ 110 K, providing unambiguous evidence for the first-order transition. The anisotropic quasiparticle relaxation establishes the three-dimensional nature of the ordered state. Below Tc, a single coherent 1.6 THz oscillation appears abruptly and remains confined to the CDW phase. This mode exhibits weak temperature dependence with negligible softening and is absent in Raman spectra. First-principles calculations imply that it is a precursor phonon at the CDW wave vector with strong electron-phonon coupling. Our results indicate that the CDW in BaFe2Al9 arises from intertwined electronic and lattice instabilities, assisted by a displacive mechanism mediated by a hidden strongly coupled phonon, distinct from conventional amplitude-mode softening scenarios.

cond-mat.str-el↗

arXiv2025

Internal multiplicity distributions of jets from nonlinear evolution within the jet function framework

Jets selected with high internal charged-particle multiplicity exhibit markedly different substructure patterns compared to inclusive jet samples. Such correlations motivate a systematic study of jet observables as a function of the normalized multiplicity, $ν= N_{\rm ch}/\langle N_{\rm ch}\rangle$. In this work, we develop a theoretical framework for the full charged-particle multiplicity distribution of exclusive and inclusive jets, formulated within the jet-function approach. The hard production and jet function are evaluated at NLO+LL$_R$ accuracy. The internal parton dynamics governing the multiplicity distribution are described by coupled nonlinear branching equations with angular ordering, supplemented by a nonperturbative modeling term that accounts for hadron-level effects. The resulting predictions are validated against \textsc{Pythia8} simulations and compared with CMS data. We examine the effects of both nonperturbative and perturbative components in shaping the multiplicity distribution, and show that Koba--Nielsen--Olesen (KNO) scaling is notably violated in the region $ν> 2$ in the full solution, with a trend consistent with Monte Carlo results. This framework that numerically solves the nonlinear multiplicity evolution goes beyond DLA-like approximations and reproduces key features seen in event generators, providing a solid foundation for future investigations of multiplicity -- conditioned jet substructure within the jet function formalism.

hep-ph↗

arXiv2025

Unlocking Multimodal Mathematical Reasoning via Process Reward Model

Process Reward Models (PRMs) have shown promise in enhancing the mathematical reasoning capabilities of Large Language Models (LLMs) through Test-Time Scaling (TTS). However, their integration into multimodal reasoning remains largely unexplored. In this work, we take the first step toward unlocking the potential of PRMs in multimodal mathematical reasoning. We identify three key challenges: (1) the scarcity of high-quality reasoning data constrains the capabilities of foundation Multimodal Large Language Models (MLLMs), which imposes further limitations on the upper bounds of TTS and reinforcement learning (RL); (2) a lack of automated methods for process labeling within multimodal contexts persists; (3) the employment of process rewards in unimodal RL faces issues like reward hacking, which may extend to multimodal scenarios. To address these issues, we introduce URSA, a three-stage Unfolding multimodal Process-Supervision Aided training framework. We first construct MMathCoT-1M, a high-quality large-scale multimodal Chain-of-Thought (CoT) reasoning dataset, to build a stronger math reasoning foundation MLLM, URSA-8B. Subsequently, we go through an automatic process to synthesize process supervision data, which emphasizes both logical correctness and perceptual consistency. We introduce DualMath-1.1M to facilitate the training of URSA-8B-RM. Finally, we propose Process-Supervised Group-Relative-Policy-Optimization (PS-GRPO), pioneering a multimodal PRM-aided online RL method that outperforms vanilla GRPO. With PS-GRPO application, URSA-8B-PS-GRPO outperforms Gemma3-12B and GPT-4o by 8.4% and 2.7% on average across 6 benchmarks. Code, data and checkpoint can be found at https://github.com/URSA-MATH.

cs.CL↗

arXiv2025

KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecasting

In the World Wide Web, reliable time series forecasts provide the forward-looking signals that drive resource planning, cache placement, and anomaly response, enabling platforms to operate efficiently as user behavior and content distributions evolve. Compared with other domains, time series forecasting for Web applications requires much faster responsiveness to support real-time decision making. We present KAIROS, a non-autoregressive time series forecasting framework that directly models segment-level multi-peak distributions. Unlike autoregressive approaches, KAIROS avoids error accumulation and achieves just-in-time inference, while improving over existing non-autoregressive models that collapse to over-smoothed predictions. Trained on the large-scale corpus, KAIROS demonstrates strong zero-shot generalization on six widely used benchmarks, delivering forecasting performance comparable to state-of-the-art foundation models with similar scale, at a fraction of their inference cost. Beyond empirical results, KAIROS highlights the importance of non-autoregressive design as a scalable paradigm for foundation models in time series.

cs.LG↗

arXiv2025

Alternating and Gaussian fermionic Isometric Tensor Network States

Isometric tensor networks in two dimensions enable efficient and accurate study of quantum many-body states, yet the effect of the isometric restriction on the represented quantum states is not fully understood. We address this question in two main contributions. First, we introduce an improved variant of isometric network states (isoTNS) in two dimensions, where the isometric arrows on the columns of the network alternate between pointing upward and downward, hence the name alternating isometric tensor network states. Second, we introduce a numerical tool -- isometric Gaussian fermionic TNS (isoGfTNS) -- that incorporates isometric constraints into the framework of Gaussian fermionic tensor network states. We demonstrate in numerous ways that alternating isoTNS represent many-body ground states of two-dimensional quantum systems significantly better than the original isoTNS. First, we show that the entanglement in an isoTNS is mediated along the isometric arrows and that alternating isoTNS mediate entanglement more efficiently than conventional isoTNS. Second, alternating isoTNS correspond to a deeper, thus more representative, sequential circuit construction of depth $O(L_x \cdot L_y)$ compared to the original isoTNS of depth $O(L_x + L_y)$. Third, using the Gaussian framework and gradient-based energy minimization, we provide numerical evidences of better bond-dimension scaling and variational energy of alternating isoGfTNS for ground states of various free fermionic models, including the Fermi surface, the band insulator, and the $p_x + ip_y$ mean-field superconductor. Finally, we find improved performance of alternating isoTNS as compared to the original isoTNS for the ground state energy of the (interacting) transverse field Ising model.

quant-ph↗

arXiv2025

LUMA: Low-Dimension Unified Motion Alignment with Dual-Path Anchoring for Text-to-Motion Diffusion Model

While current diffusion-based models, typically built on U-Net architectures, have shown promising results on the text-to-motion generation task, they still suffer from semantic misalignment and kinematic artifacts. Through analysis, we identify severe gradient attenuation in the deep layers of the network as a key bottleneck, leading to insufficient learning of high-level features. To address this issue, we propose \textbf{LUMA} (\textit{\textbf{L}ow-dimension \textbf{U}nified \textbf{M}otion \textbf{A}lignment}), a text-to-motion diffusion model that incorporates dual-path anchoring to enhance semantic alignment. The first path incorporates a lightweight MoCLIP model trained via contrastive learning without relying on external data, offering semantic supervision in the temporal domain. The second path introduces complementary alignment signals in the frequency domain, extracted from low-frequency DCT components known for their rich semantic content. These two anchors are adaptively fused through a temporal modulation mechanism, allowing the model to progressively transition from coarse alignment to fine-grained semantic refinement throughout the denoising process. Experimental results on HumanML3D and KIT-ML demonstrate that LUMA achieves state-of-the-art performance, with FID scores of 0.035 and 0.123, respectively. Furthermore, LUMA accelerates convergence by 1.4$\times$ compared to the baseline, making it an efficient and scalable solution for high-fidelity text-to-motion generation.

cs.CV↗

arXiv2025

Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think

REPA and its variants effectively mitigate training challenges in diffusion models by incorporating external visual representations from pretrained models, through alignment between the noisy hidden projections of denoising networks and foundational clean image representations. We argue that the external alignment, which is absent during the entire denoising inference process, falls short of fully harnessing the potential of discriminative representations. In this work, we propose a straightforward method called Representation Entanglement for Generation (REG), which entangles low-level image latents with a single high-level class token from pretrained foundation models for denoising. REG acquires the capability to produce coherent image-class pairs directly from pure noise, substantially improving both generation quality and training efficiency. This is accomplished with negligible additional inference overhead, requiring only one single additional token for denoising (<0.5\% increase in FLOPs and latency). The inference process concurrently reconstructs both image latents and their corresponding global semantics, where the acquired semantic knowledge actively guides and enhances the image generation process. On ImageNet 256$\times$256, SiT-XL/2 + REG demonstrates remarkable convergence acceleration, achieving $\textbf{63}\times$ and $\textbf{23}\times$ faster training than SiT-XL/2 and SiT-XL/2 + REPA, respectively. More impressively, SiT-L/2 + REG trained for merely 400K iterations outperforms SiT-XL/2 + REPA trained for 4M iterations ($\textbf{10}\times$ longer). Code is available at: https://github.com/Martinser/REG.

cs.CV↗

arXiv2025

Losing is for Cherishing: Data Valuation Based on Machine Unlearning and Shapley Value

The proliferation of large models has intensified the need for efficient data valuation methods to quantify the contribution of individual data providers. Traditional approaches, such as game-theory-based Shapley value and influence-function-based techniques, face prohibitive computational costs or require access to full data and model training details, making them hardly achieve partial data valuation. To address this, we propose Unlearning Shapley, a novel framework that leverages machine unlearning to estimate data values efficiently. By unlearning target data from a pretrained model and measuring performance shifts on a reachable test set, our method computes Shapley values via Monte Carlo sampling, avoiding retraining and eliminating dependence on full data. Crucially, Unlearning Shapley supports both full and partial data valuation, making it scalable for large models (e.g., LLMs) and practical for data markets. Experiments on benchmark datasets and large-scale text corpora demonstrate that our approach matches the accuracy of state-of-the-art methods while reducing computational overhead by orders of magnitude. Further analysis confirms a strong correlation between estimated values and the true impact of data subsets, validating its reliability in real-world scenarios. This work bridges the gap between data valuation theory and practical deployment, offering a scalable, privacy-compliant solution for modern AI ecosystems.

cs.AI↗

arXiv2025

TableEval: A Real-World Benchmark for Complex, Multilingual, and Multi-Structured Table Question Answering

LLMs have shown impressive progress in natural language processing. However, they still face significant challenges in TableQA, where real-world complexities such as diverse table structures, multilingual data, and domain-specific reasoning are crucial. Existing TableQA benchmarks are often limited by their focus on simple flat tables and suffer from data leakage. Furthermore, most benchmarks are monolingual and fail to capture the cross-lingual and cross-domain variability in practical applications. To address these limitations, we introduce TableEval, a new benchmark designed to evaluate LLMs on realistic TableQA tasks. Specifically, TableEval includes tables with various structures (such as concise, hierarchical, and nested tables) collected from four domains (including government, finance, academia, and industry reports). Besides, TableEval features cross-lingual scenarios with tables in Simplified Chinese, Traditional Chinese, and English. To minimize the risk of data leakage, we collect all data from recent real-world documents. Considering that existing TableQA metrics fail to capture semantic accuracy, we further propose SEAT, a new evaluation framework that assesses the alignment between model responses and reference answers at the sub-question level. Experimental results have shown that SEAT achieves high agreement with human judgment. Extensive experiments on TableEval reveal critical gaps in the ability of state-of-the-art LLMs to handle these complex, real-world TableQA tasks, offering insights for future improvements. We make our dataset available here: https://github.com/wenge-research/TableEval.

cs.CL↗

arXiv2025

FedEL: Federated Elastic Learning for Heterogeneous Devices

Federated learning (FL) enables distributed devices to collaboratively train machine learning models while maintaining data privacy. However, the heterogeneous hardware capabilities of devices often result in significant training delays, as straggler clients with limited resources prolong the aggregation process. Existing solutions such as client selection, asynchronous FL, and partial training partially address these challenges but encounter issues such as reduced accuracy, stale updates, and compromised model performance due to inconsistent training contributions. To overcome these limitations, we propose FedEL, a federated elastic learning framework that enhances training efficiency while maintaining model accuracy. FedEL introduces a novel window-based training process, sliding the window to locate the training part of the model and dynamically selecting important tensors for training within a coordinated runtime budget. This approach ensures progressive and balanced training across all clients, including stragglers. Additionally, FedEL employs a tensor importance adjustment module, harmonizing local and global tensor importance to mitigate biases caused by data heterogeneity. The experiment results show that FedEL achieves up to 3.87x improvement in time-to-accuracy compared to baselines while maintaining or exceeding final test accuracy.

cs.LG↗

arXiv2025

Improving Anomalous Sound Detection with Attribute-aware Representation from Domain-adaptive Pre-training

Anomalous Sound Detection (ASD) is often formulated as a machine attribute classification task, a strategy necessitated by the common scenario where only normal data is available for training. However, the exhaustive collection of machine attribute labels is laborious and impractical. To address the challenge of missing attribute labels, this paper proposes an agglomerative hierarchical clustering method for the assignment of pseudo-attribute labels using representations derived from a domain-adaptive pre-trained model, which are expected to capture machine attribute characteristics. We then apply model adaptation to this pre-trained model through supervised fine-tuning for machine attribute classification, resulting in a new state-of-the-art performance. Evaluation on the Detection and Classification of Acoustic Scenes and Events (DCASE) 2025 Challenge dataset demonstrates that our proposed approach yields significant performance gains, ultimately outperforming our previous top-ranking system in the challenge.

cs.SD↗

arXiv2025

Meow: End-to-End Outline Writing for Automatic Academic Survey

As academic paper publication numbers grow exponentially, conducting in-depth surveys with LLMs automatically has become an inevitable trend. Outline writing, which aims to systematically organize related works, is critical for automated survey generation. Yet existing automatic survey methods treat outline writing as mere workflow steps in the overall pipeline. Such template-based workflows produce outlines that lack in-depth understanding of the survey topic and fine-grained styles. To address these limitations, we propose Meow, the first metadata-driven outline writing framework that produces organized and faithful outlines efficiently. Specifically, we first formulate outline writing as an end-to-end task that generates hierarchical structured outlines from paper metadata. We then curate a high-quality dataset of surveys from arXiv, bioRxiv, and medRxiv, and establish systematic evaluation metrics for outline quality assessment. Finally, we employ a two-stage training approach combining supervised fine-tuning and reinforcement learning. Our 8B reasoning model demonstrates strong performance with high structural fidelity and stylistic coherence.

cs.CL↗