arXiv subjects

Pengyu Wang

Publications and source records attributed to Pengyu Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

NEFT: A Unified Transformer Framework for Efficient Near-Field CSI Feedback in XL-MIMO Systems

Extremely large-scale multiple-input multiple-output (XL-MIMO) systems, operating in the near-field region due to their massive antenna arrays, are key enablers of next-generation wireless communications but face significant challenges in channel state information (CSI) feedback. Deep learning has emerged as a powerful tool by learning compact channel features for feedback. However, existing methods struggle to capture the intricate structure of near-field CSI and incur prohibitive computational overhead on practical mobile devices. To overcome these limitations, we propose the near-field efficient feedback Transformer (NEFT) family for accurate near-field CSI feedback with reduced overhead under diverse hardware constraints. NEFT builds on a hierarchical vision Transformer backbone with progressive token reduction and multi-scale feature extraction, enabling compact and effective modeling of near-field channel characteristics. Furthermore, NEFT is extended with lightweight variants: NEFT-Compact applies multi-level knowledge distillation (KD) to reduce model complexity while preserving accuracy; NEFT-Hybrid adopts an attention-free CNN encoder to reduce encoder-side computation; and NEFT-Edge combines NEFT-Hybrid with KD to enable deployment on highly resource-constrained edge devices. Extensive simulations show that NEFT achieves a 15--21dB improvement in normalized mean-squared error over state-of-the-art methods, NEFT-Compact and NEFT-Edge reduce total FLOPs by 25-36% with negligible accuracy loss, while NEFT-Hybrid reduces encoder-side complexity by up to 64%, enabling deployment in highly asymmetric device scenarios. These results establish NEFT as a practical and scalable solution for near-field CSI feedback in XL-MIMO systems.

eess.SP↗

arXiv2025

LoPT: Lossless Parallel Tokenization Acceleration for Long Context Inference of Large Language Model

Long context inference scenarios have become increasingly important for large language models, yet they introduce significant computational latency. While prior research has optimized long-sequence inference through operators, model architectures, and system frameworks, tokenization remains an overlooked bottleneck. Existing parallel tokenization methods accelerate processing through text segmentation and multi-process tokenization, but they suffer from inconsistent results due to boundary artifacts that occur after merging. To address this, we propose LoPT, a novel Lossless Parallel Tokenization framework that ensures output identical to standard sequential tokenization. Our approach employs character-position-based matching and dynamic chunk length adjustment to align and merge tokenized segments accurately. Extensive experiments across diverse long-text datasets demonstrate that LoPT achieves significant speedup while guaranteeing lossless tokenization. We also provide theoretical proof of consistency and comprehensive analytical studies to validate the robustness of our method.

cs.CL↗

arXiv2025

MS-BART: Unified Modeling of Mass Spectra and Molecules for Structure Elucidation

Mass spectrometry (MS) plays a critical role in molecular identification, significantly advancing scientific discovery. However, structure elucidation from MS data remains challenging due to the scarcity of annotated spectra. While large-scale pretraining has proven effective in addressing data scarcity in other domains, applying this paradigm to mass spectrometry is hindered by the complexity and heterogeneity of raw spectral signals. To address this, we propose MS-BART, a unified modeling framework that maps mass spectra and molecular structures into a shared token vocabulary, enabling cross-modal learning through large-scale pretraining on reliably computed fingerprint-molecule datasets. Multi-task pretraining objectives further enhance MS-BART's generalization by jointly optimizing denoising and translation task. The pretrained model is subsequently transferred to experimental spectra through finetuning on fingerprint predictions generated with MIST, a pre-trained spectral inference model, thereby enhancing robustness to real-world spectral variability. While finetuning alleviates the distributional difference, MS-BART still suffers molecular hallucination and requires further alignment. We therefore introduce a chemical feedback mechanism that guides the model toward generating molecules closer to the reference structure. Extensive evaluations demonstrate that MS-BART achieves SOTA performance across 5/12 key metrics on MassSpecGym and NPLIB1 and is faster by one order of magnitude than competing diffusion-based methods, while comprehensive ablation studies systematically validate the model's effectiveness and robustness.

cs.LG↗

arXiv2025

FedAGHN: Personalized Federated Learning with Attentive Graph HyperNetworks

Personalized Federated Learning (PFL) aims to address the statistical heterogeneity of data across clients by learning the personalized model for each client. Among various PFL approaches, the personalized aggregation-based approach conducts parameter aggregation in the server-side aggregation phase to generate personalized models, and focuses on learning appropriate collaborative relationships among clients for aggregation. However, the collaborative relationships vary in different scenarios and even at different stages of the FL process. To this end, we propose Personalized Federated Learning with Attentive Graph HyperNetworks (FedAGHN), which employs Attentive Graph HyperNetworks (AGHNs) to dynamically capture fine-grained collaborative relationships and generate client-specific personalized initial models. Specifically, AGHNs empower graphs to explicitly model the client-specific collaborative relationships, construct collaboration graphs, and introduce tunable attentive mechanism to derive the collaboration weights, so that the personalized initial models can be obtained by aggregating parameters over the collaboration graphs. Extensive experiments can demonstrate the superiority of FedAGHN. Moreover, a series of visualizations are presented to explore the effectiveness of collaboration graphs learned by FedAGHN.

cs.LG↗

arXiv2025

Normal mode parameters estimation by a VLA in single-shooting

This paper proposes an orthogonality-constrained modal search (OCMS) method for estimating modal wavenumbers and modal depth functions using a vertical linear array (VLA). Under the assumption of a known sound speed profile, OCMS leverages the orthogonality of distinct modal depth functions to extract both the modal depth functions and their corresponding wavenumbers, even when the VLA and a monochromatic sound source remain stationary.The performance of OCMS is evaluated through numerical simulations under varying signal-to-noise ratios (SNRs), different VLA apertures, varying numbers of VLA elements, VLA tilt and sound speed profile (SSP) uncertainty. The results demonstrate that OCMS is robust against noise, VLA aperture variations, and changes in the number of VLA elements, meanwhile, the algorithm maintains reliable performance when SSP uncertainty < 1 m/s and VLA tilt angle <5°. Furthermore, the effectiveness of OCMS is validated using SwellEx96 experimental data. The relative error between the modal wavenumbers derived from experimental data and those computed via Kraken is on the order of $10^{-4}$.

eess.SP↗

arXiv2025

Decoupled Proxy Alignment: Mitigating Language Prior Conflict for Multimodal Alignment in MLLM

Multimodal large language models (MLLMs) have gained significant attention due to their impressive ability to integrate vision and language modalities. Recent advancements in MLLMs have primarily focused on improving performance through high-quality datasets, novel architectures, and optimized training strategies. However, in this paper, we identify a previously overlooked issue, language prior conflict, a mismatch between the inherent language priors of large language models (LLMs) and the language priors in training datasets. This conflict leads to suboptimal vision-language alignment, as MLLMs are prone to adapting to the language style of training samples. To address this issue, we propose a novel training method called Decoupled Proxy Alignment (DPA). DPA introduces two key innovations: (1) the use of a proxy LLM during pretraining to decouple the vision-language alignment process from language prior interference, and (2) dynamic loss adjustment based on visual relevance to strengthen optimization signals for visually relevant tokens. Extensive experiments demonstrate that DPA significantly mitigates the language prior conflict, achieving superior alignment performance across diverse datasets, model families, and scales. Our method not only improves the effectiveness of MLLM training but also shows exceptional generalization capabilities, making it a robust approach for vision-language alignment. Our code is available at https://github.com/fnlp-vision/DPA.

cs.CL↗

arXiv2025

UnifiedVisual: A Framework for Constructing Unified Vision-Language Datasets

Unified vision large language models (VLLMs) have recently achieved impressive advancements in both multimodal understanding and generation, powering applications such as visual question answering and text-guided image synthesis. However, progress in unified VLLMs remains constrained by the lack of datasets that fully exploit the synergistic potential between these two core abilities. Existing datasets typically address understanding and generation in isolation, thereby limiting the performance of unified VLLMs. To bridge this critical gap, we introduce a novel dataset construction framework, UnifiedVisual, and present UnifiedVisual-240K, a high-quality dataset meticulously designed to facilitate mutual enhancement between multimodal understanding and generation. UnifiedVisual-240K seamlessly integrates diverse visual and textual inputs and outputs, enabling comprehensive cross-modal reasoning and precise text-to-image alignment. Our dataset encompasses a wide spectrum of tasks and data sources, ensuring rich diversity and addressing key shortcomings of prior resources. Extensive experiments demonstrate that models trained on UnifiedVisual-240K consistently achieve strong performance across a wide range of tasks. Notably, these models exhibit significant mutual reinforcement between multimodal understanding and generation, further validating the effectiveness of our framework and dataset. We believe UnifiedVisual represents a new growth point for advancing unified VLLMs and unlocking their full potential. Our code and datasets is available at https://github.com/fnlp-vision/UnifiedVisual.

cs.CL↗

arXiv2025

$L^2$-solutions to stochastic reaction-diffusion equations with superlinear drifts driven by space-time white noise^

Consider the following stochastic reaction-diffusion equation with logarithmic superlinear coefficient b, driven by space-time white noise W: $$ u_t(t,x) = (1/2)u_{xx}(t,x) + b(u(t,x)) + σ(u(t,x))W(dt,dx) $$ for $t > 0$ and $x \in [0,1]$, with initial condition $$ u(0,x) = u_0(x) $$ for $x \in [0,1]$, where $u_0 \in L^2[0,1]$. In this paper, we establish existence and uniqueness of probabilistically strong solutions in $C(R_+, L^2[0,1])$. Our result resolves a problem from [Ann. Probab. 47 (2019) 519-559] and provides an alternative proof of the non-blowup of $L^2[0,1]$ solutions from the same reference. We use new Gronwall-type inequalities. Due to nonlinearity, we work with first order moments, requiring precise estimates of the stochastic convolution.

math.PR↗

arXiv2025

VINP: Variational Bayesian Inference with Neural Speech Prior for Joint ASR-Effective Speech Dereverberation and Blind RIR Identification

Reverberant speech, denoting the speech signal degraded by reverberation, contains crucial knowledge of both anechoic source speech and room impulse response (RIR). This work proposes a variational Bayesian inference (VBI) framework with neural speech prior (VINP) for joint speech dereverberation and blind RIR identification. In VINP, a probabilistic signal model is constructed in the time-frequency (T-F) domain based on convolution transfer function (CTF) approximation. For the first time, we propose using an arbitrary discriminative dereverberation deep neural network (DNN) to estimate the prior distribution of anechoic speech within a probabilistic model. By integrating both reverberant speech and the anechoic speech prior, VINP yields the maximum a posteriori (MAP) and maximum likelihood (ML) estimations of the anechoic speech spectrum and CTF filter, respectively. After simple transformations, the waveforms of anechoic speech and RIR are estimated. VINP is effective for automatic speech recognition (ASR) systems, which sets it apart from most deep learning (DL)-based single-channel dereverberation approaches. Experiments on single-channel speech dereverberation demonstrate that VINP attains state-of-the-art (SOTA) performance in mean opinion score (MOS) and word error rate (WER). For blind RIR identification, experiments demonstrate that VINP achieves SOTA performance in estimating reverberation time at 60 dB (RT60) and advanced performance in direct-to-reverberation ratio (DRR) estimation. Codes and audio samples are available online.

eess.AS↗

arXiv2025

Jamming Identification with Differential Transformer for Low-Altitude Wireless Networks

Wireless jamming identification, which detects and classifies electromagnetic jamming from non-cooperative devices, is crucial for emerging low-altitude wireless networks consisting of many drone terminals that are highly susceptible to electromagnetic jamming. However, jamming identification schemes adopting deep learning (DL) are vulnerable to attacks involving carefully crafted adversarial samples, resulting in inevitable robustness degradation. To address this issue, we propose a differential transformer framework for wireless jamming identification. Firstly, we introduce a differential transformer network in order to distinguish jamming signals, which overcomes the attention noise when compared with its traditional counterpart by performing self-attention operations in a differential manner. Secondly, we propose a randomized masking training strategy to improve network robustness, which leverages the patch partitioning mechanism inherent to transformer architectures in order to create parallel feature extraction branches. Each branch operates on a distinct, randomly masked subset of patches, which fundamentally constrains the propagation of adversarial perturbations across the network. Additionally, the ensemble effect generated by fusing predictions from these diverse branches demonstrates superior resilience against adversarial attacks. Finally, we introduce a novel consistent training framework that significantly enhances adversarial robustness through dualbranch regularization. Simulation results demonstrate that our proposed methodology is superior to existing methods in boosting robustness to adversarial samples.

eess.SP↗

arXiv2025

Interior-Point-based H2 Controller Synthesis for Compartmental Systems

This paper addresses the problem of the optimal $H_2$ controller design for compartmental systems. In other words, we aim to enhance system robustness while maintaining the law of mass conservation. We perform a novel problem transformation and establish that the original problem is equivalent to an new optimization problem with a closed polyhedron constraint. Existing works have developed various first-order methods to tackle inequality constraints. However, the performance of the first-order method is limited in terms of convergence speed and precision, restricting its potential in practical applications. Therefore, developing a novel algorithm with fast speed and high precision is critical. In this paper, we reformulate the problem using log-barrier functions and introduce two separate approaches to address the problem: the first-order interior point method (FIPM) and the second-order interior point method (SIPM). We show they converge to a stationary point of the new problem. In addition, we propose an initialization method to guarantee the interior property of initial values. Finally, we compare FIPM and SIPM through a room temperature control example and show their pros and cons.

math.OC↗

arXiv2025

Proactive Constrained Policy Optimization with Preemptive Penalty

Safe Reinforcement Learning (RL) often faces significant issues such as constraint violations and instability, necessitating the use of constrained policy optimization, which seeks optimal policies while ensuring adherence to specific constraints like safety. Typically, constrained optimization problems are addressed by the Lagrangian method, a post-violation remedial approach that may result in oscillations and overshoots. Motivated by this, we propose a novel method named Proactive Constrained Policy Optimization (PCPO) that incorporates a preemptive penalty mechanism. This mechanism integrates barrier items into the objective function as the policy nears the boundary, imposing a cost. Meanwhile, we introduce a constraint-aware intrinsic reward to guide boundary-aware exploration, which is activated only when the policy approaches the constraint boundary. We establish theoretical upper and lower bounds for the duality gap and the performance of the PCPO update, shedding light on the method's convergence characteristics. Additionally, to enhance the optimization performance, we adopt a policy iteration approach. An interesting finding is that PCPO demonstrates significant stability in experiments. Experimental results indicate that the PCPO framework provides a robust solution for policy optimization under constraints, with important implications for future research and practical applications.

cs.LG↗

arXiv2025

CleanMel: Mel-Spectrogram Enhancement for Improving Both Speech Quality and ASR

In this work, we propose CleanMel, a single-channel Mel-spectrogram denoising and dereverberation network for improving both speech quality and automatic speech recognition (ASR) performance. The proposed network takes as input the noisy and reverberant microphone recording and predicts the corresponding clean Mel-spectrogram. The enhanced Mel-spectrogram can be either transformed to the speech waveform with a neural vocoder or directly used for ASR. The proposed network is composed of interleaved cross-band and narrow-band processing in the Mel-frequency domain, for learning the full-band spectral pattern and the narrow-band properties of signals, respectively. Compared to linear-frequency domain or time-domain speech enhancement, the key advantage of Mel-spectrogram enhancement is that Mel-frequency presents speech in a more compact way and thus is easier to learn, which will benefit both speech quality and ASR. Experimental results on five English and one Chinese datasets demonstrate a significant improvement in both speech quality and ASR performance achieved by the proposed model.Code and audio examples of our model are available online.

eess.AS↗

arXiv2025

EBaReT: Expert-guided Bag Reward Transformer for Auto Bidding

Reinforcement learning has been widely applied in automated bidding. Traditional approaches model bidding as a Markov Decision Process (MDP). Recently, some studies have explored using generative reinforcement learning methods to address long-term dependency issues in bidding environments. Although effective, these methods typically rely on supervised learning approaches, which are vulnerable to low data quality due to the amount of sub-optimal bids and low probability rewards resulting from the low click and conversion rates. Unfortunately, few studies have addressed these challenges. In this paper, we formalize the automated bidding as a sequence decision-making problem and propose a novel Expert-guided Bag Reward Transformer (EBaReT) to address concerns related to data quality and uncertainty rewards. Specifically, to tackle data quality issues, we generate a set of expert trajectories to serve as supplementary data in the training process and employ a Positive-Unlabeled (PU) learning-based discriminator to identify expert transitions. To ensure the decision also meets the expert level, we further design a novel expert-guided inference strategy. Moreover, to mitigate the uncertainty of rewards, we consider the transitions within a certain period as a "bag" and carefully design a reward function that leads to a smoother acquisition of rewards. Extensive experiments demonstrate that our model achieves superior performance compared to state-of-the-art bidding methods.

cs.LG↗

arXiv2025

Adversarial Preference Learning for Robust LLM Alignment

Modern language models often rely on Reinforcement Learning from Human Feedback (RLHF) to encourage safe behaviors. However, they remain vulnerable to adversarial attacks due to three key limitations: (1) the inefficiency and high cost of human annotation, (2) the vast diversity of potential adversarial attacks, and (3) the risk of feedback bias and reward hacking. To address these challenges, we introduce Adversarial Preference Learning (APL), an iterative adversarial training method incorporating three key innovations. First, a direct harmfulness metric based on the model's intrinsic preference probabilities, eliminating reliance on external assessment. Second, a conditional generative attacker that synthesizes input-specific adversarial variations. Third, an iterative framework with automated closed-loop feedback, enabling continuous adaptation through vulnerability discovery and mitigation. Experiments on Mistral-7B-Instruct-v0.3 demonstrate that APL significantly enhances robustness, achieving 83.33% harmlessness win rate over the base model (evaluated by GPT-4o), reducing harmful outputs from 5.88% to 0.43% (measured by LLaMA-Guard), and lowering attack success rate by up to 65% according to HarmBench. Notably, APL maintains competitive utility, with an MT-Bench score of 6.59 (comparable to the baseline 6.78) and an LC-WinRate of 46.52% against the base model.

cs.LG↗

arXiv2025

MRGAgents: A Multi-Agent Framework for Improved Medical Report Generation with Med-LVLMs

Medical Large Vision-Language Models (Med-LVLMs) have been widely adopted for medical report generation. Despite Med-LVLMs producing state-of-the-art performance, they exhibit a bias toward predicting all findings as normal, leading to reports that overlook critical abnormalities. Furthermore, these models often fail to provide comprehensive descriptions of radiologically relevant regions necessary for accurate diagnosis. To address these challenges, we proposeMedical Report Generation Agents (MRGAgents), a novel multi-agent framework that fine-tunes specialized agents for different disease categories. By curating subsets of the IU X-ray and MIMIC-CXR datasets to train disease-specific agents, MRGAgents generates reports that more effectively balance normal and abnormal findings while ensuring a comprehensive description of clinically relevant regions. Our experiments demonstrate that MRGAgents outperformed the state-of-the-art, improving both report comprehensiveness and diagnostic utility.

cs.MA↗

arXiv2025

PierGuard: A Planning Framework for Underwater Robotic Inspection of Coastal Piers

Using underwater robots instead of humans for the inspection of coastal piers can enhance efficiency while reducing risks. A key challenge in performing these tasks lies in achieving efficient and rapid path planning within complex environments. Sampling-based path planning methods, such as Rapidly-exploring Random Tree* (RRT*), have demonstrated notable performance in high-dimensional spaces. In recent years, researchers have begun designing various geometry-inspired heuristics and neural network-driven heuristics to further enhance the effectiveness of RRT*. However, the performance of these general path planning methods still requires improvement when applied to highly cluttered underwater environments. In this paper, we propose PierGuard, which combines the strengths of bidirectional search and neural network-driven heuristic regions. We design a specialized neural network to generate high-quality heuristic regions in cluttered maps, thereby improving the performance of the path planning. Through extensive simulation and real-world ocean field experiments, we demonstrate the effectiveness and efficiency of our proposed method compared with previous research. Our method achieves approximately 2.6 times the performance of the state-of-the-art geometric-based sampling method and nearly 4.9 times that of the state-of-the-art learning-based sampling method. Our results provide valuable insights for the automation of pier inspection and the enhancement of maritime safety. The updated experimental video is available in the supplementary materials.

cs.RO↗

arXiv2025

Sparse $H_\infty$ Controller for Networked Control Systems: Non-Structured and Optimal Structured Design

This paper provides a comprehensive analysis of the design of optimal structured and sparse $H_\infty$ controllers for continuous-time linear time-invariant (LTI) systems. Three problems are considered. First, designing the sparsest $H_\infty$ controller, which minimizes the sparsity of the controller while satisfying the given performance requirements. Second, designing a sparsity-promoting $H_\infty$ controller, which balances system performance and controller sparsity. Third, designing a $H_\infty$ controller subject to a structural constraint, which enhances system performance with a specified sparsity pattern. For each problem, we adopt a linearization technique that transforms the original nonconvex problem into a convex semidefinite programming (SDP) problem. Subsequently, we design an iterative linear matrix inequality (ILMI) algorithm for each problem, which ensures guaranteed convergence. We further characterize the first-order optimality using the Karush-Kuhn-Tucker (KKT) conditions and prove that any limit point of the solution sequence generated by the ILMI algorithm is a stationary point. For the first and second problems, we validate that our algorithms can reduce the number of non-zero elements and thus the communication burden through several numerical simulations. For the third problem, we refine the solutions obtained in existing literature, demonstrating that our approaches achieve significant improvements.

math.OC↗