arXiv subjects

Zhu Meng

Publications and source records attributed to Zhu Meng.

15 recordsLinked to original sources

arXiv2026

Study the Longitudinal Entropy Deposition using d+Au Collision

Relativistic hydrodynamics successfully describes bulk observables in symmetric heavy-ion collisions, but struggles to reproduce charged-particle rapidity distributions in asymmetric systems such as d+Au collisions. To address this challenge, we introduce two key improvements to the initial-state modeling: sampling deuteron configurations from an ab initio wavefunction, and developing a new longitudinal entropy deposition model that incorporates a transverse entropy deposition coefficient $\beta$ and a rapidity loss term scaling with the number of binary collisions $n_{\rm BC}$. Using the (3+1)-dimensional viscous hydrodynamic model CLVisc coupled with the SMASH afterburner, we simulate d+Au collisions at $\sqrt{s_{\rm NN}} = 200$ GeV and successfully reproduce the experimental charged-particle pseudorapidity distributions across five centrality classes with $\beta = 0.35$, as well as the transverse momentum spectra and anisotropic flow $v_n$. The entropy deposition coefficient $\beta$ and the $n_{\rm BC}$-dependent rapidity loss are found to play crucial roles in achieving this agreement. Furthermore, this longitudinal entropy deposition framework demonstrates excellent universality, as validated in p+Au, $^3$He+Au, and Au+Au collisions. Our entropy deposition mechanism could be widely applied to recent light-nucleus collisions such as O+O, Ne+Ne, and asymmetric systems like Pb+Ne at LHC energies, thereby better constraining the nuclear structure of light nuclei through an improved longitudinal description.

nucl-th↗

arXiv2026

ToolPrivacyBench: Benchmarking Purpose-Bound Privacy in Tool-Using LLM Agents

Large language models (LLMs) have increasingly moved from standalone text generation systems to agents that invoke external tools, access environments, and execute multi-step tasks. However, conventional function-calling benchmarks mainly evaluate task completion and API correctness, while privacy evaluation benchmarks typically focus on final responses or privacy judgments. Neither perspective captures purpose-bound information flow across an executed multi-tool trajectory. Motivated by this limitation in current agent evaluation, ToolPrivacyBench audits whether task-private atoms are routed only to authorized tools and downstream sinks, thereby evaluating both task completion and privacy over-disclosure during tool use. The benchmark contains 2,150 cases, including 1,150 fully synthetic privacy-sensitive business workflows and 1,000 cases adapted from existing multi-tool and function-calling benchmarks. Each case is represented by a policy knowledge base. After an agent executes against mock business backends, the evaluator compares recorded tool arguments and backend audit logs with this policy knowledge base. The evaluation covers nine widely used agents to characterize purpose-bound privacy over-disclosure. The results show that successful tool execution does not imply appropriate privacy disclosure: an agent may complete a task while transmitting unnecessary private information through intermediate tool calls. ToolPrivacyBench therefore formalizes a need-to-know disclosure boundary, under which each tool should receive only the information necessary for its stated purpose, and uses trajectory-level auditing to identify privacy over-disclosure in multi-tool workflows.

cs.CR↗

arXiv2026

CHRep: Cross-modal Histology Representation and Post-hoc Calibration for Spatial Gene Expression Prediction

Spatial transcriptomics (ST) enables spatially resolved gene profiling but remains expensive and low-throughput, limiting large-cohort studies and routine clinical use. Predicting spatial gene expression from routine hematoxylin and eosin (H&E) slides is a promising alternative, yet under realistic leave-one-slide-out evaluation, existing models often suffer from slide-level appearance shifts and regression-driven over-smoothing that suppress biologically meaningful variation. CHRep is a two-phase framework for robust histology-to-expression prediction. In the training phase, CHRep learns a structure-aware representation by jointly optimizing correlation-aware regression, symmetric image-expression alignment, and coordinate-induced spatial topology regularization. In the inference phase, cross-slide robustness is improved without backbone fine-tuning through a lightweight calibration module trained on the training slides, which combines a non-parametric estimate from a training gallery with a magnitude-regularized correction module. Unlike prior embedding-alignment or retrieval-based transfer methods that rely on a single prediction route, CHRep couples topology-preserving representation learning with post-hoc calibration, enabling stable neighborhood retrieval and controlled bias correction under slide-level shifts. Across the three cohorts, CHRep consistently improves gene-wise correlation under leave-one-slide-out evaluation, with the largest gains observed on Alex+10x. Relative to HAGE, the Pearson correlation coefficient on all considered genes [PCC(ACG)] increases by 4.0% on cSCC and 9.8% on HER2+. Relative to mclSTExp, PCC(ACG) further improves by 39.5% on Alex+10x, together with 9.7% and 9.0% reductions in mean squared error (MSE) and mean absolute error (MAE), respectively.

cs.CV↗

arXiv2026

PathGLS: Evaluating Pathology Vision-Language Models without Ground Truth through Multi-Dimensional Consistency

Vision-Language Models (VLMs) offer significant potential in computational pathology by enabling interpretable image analysis, automated reporting, and scalable decision support. However, their widespread clinical adoption remains limited due to the absence of reliable, automated evaluation metrics capable of identifying subtle failures such as hallucinations. To address this gap, we propose PathGLS, a novel reference-free evaluation framework that assesses pathology VLMs across three dimensions: Grounding (fine-grained visual-text alignment), Logic (entailment graph consistency using Natural Language Inference), and Stability (output variance under adversarial visual-semantic perturbations). PathGLS supports both patch-level and whole-slide image (WSI)-level analysis, yielding a comprehensive trust score. Experiments on Quilt-1M, TCGA, REG2025, PathMMU and TCGA-Sarcoma datasets demonstrate the superiority of PathGLS. Specifically, on the Quilt-1M dataset, PathGLS reveals a steep sensitivity drop of 40.2% for hallucinated reports compared to only 2.1% for BERTScore. Moreover, validation against expert-defined clinical error hierarchies reveals that PathGLS achieves a strong Spearman's rank correlation of $\rho=0.71$ ($p < 0.0001$), significantly outperforming Large Language Model (LLM)-based approaches (Gemini 3.0 Pro: $\rho=0.39$, $p < 0.0001$). These results establish PathGLS as a robust reference-free metric. By directly quantifying hallucination rates and domain shift robustness, it serves as a reliable criterion for benchmarking VLMs on private clinical datasets and informing safe deployment. Code can be found at: https://github.com/My13ad/PathGLS

cs.CV↗

arXiv2025

Post-TIPS Prediction via Multimodal Interaction: A Multi-Center Dataset and Framework for Survival, Complication, and Portal Pressure Assessment

Transjugular intrahepatic portosystemic shunt (TIPS) is an established procedure for portal hypertension, but provides variable survival outcomes and frequent overt hepatic encephalopathy (OHE), indicating the necessity of accurate preoperative prognostic modeling. Current studies typically build machine learning models from preoperative CT images or clinical characteristics, but face three key challenges: (1) labor-intensive region-of-interest (ROI) annotation, (2) poor reliability and generalizability of unimodal methods, and (3) incomplete assessment from single-endpoint prediction. Moreover, the lack of publicly accessible datasets constrains research in this field. Therefore, we present MultiTIPS, the first public multi-center dataset for TIPS prognosis, and propose a novel multimodal prognostic framework based on it. The framework comprises three core modules: (1) dual-option segmentation, which integrates semi-supervised and foundation model-based pipelines to achieve robust ROI segmentation with limited annotations and facilitate subsequent feature extraction; (2) multimodal interaction, where three techniques, multi-grained radiomics attention (MGRA), progressive orthogonal disentanglement (POD), and clinically guided prognostic enhancement (CGPE), are introduced to enable cross-modal feature interaction and complementary representation integration, thus improving model accuracy and robustness; and (3) multi-task prediction, where a staged training strategy is used to perform stable optimization of survival, portal pressure gradient (PPG), and OHE prediction for comprehensive prognostic assessment. Extensive experiments on MultiTIPS demonstrate the superiority of the proposed method over state-of-the-art approaches, along with strong cross-domain generalization and interpretability, indicating its promise for clinical application. The dataset and code are available.

cs.CV↗

arXiv2025

OLMA: One Loss for More Accurate Time Series Forecasting

Time series forecasting faces two important but often overlooked challenges. Firstly, the inherent random noise in the time series labels sets a theoretical lower bound for the forecasting error, which is positively correlated with the entropy of the labels. Secondly, neural networks exhibit a frequency bias when modeling the state-space of time series, that is, the model performs well in learning certain frequency bands but poorly in others, thus restricting the overall forecasting performance. To address the first challenge, we prove a theorem that there exists a unitary transformation that can reduce the marginal entropy of multiple correlated Gaussian processes, thereby providing guidance for reducing the lower bound of forecasting error. Furthermore, experiments confirm that Discrete Fourier Transform (DFT) can reduce the entropy in the majority of scenarios. Correspondingly, to alleviate the frequency bias, we jointly introduce supervision in the frequency domain along the temporal dimension through DFT and Discrete Wavelet Transform (DWT). This supervision-side strategy is highly general and can be seamlessly integrated into any supervised learning method. Moreover, we propose a novel loss function named OLMA, which utilizes the frequency domain transformation across both channel and temporal dimensions to enhance forecasting. Finally, the experimental results on multiple datasets demonstrate the effectiveness of OLMA in addressing the above two challenges and the resulting improvement in forecasting accuracy. The results also indicate that the perspectives of entropy and frequency bias provide a new and feasible research direction for time series forecasting. The code is available at: https://github.com/Yuyun1011/OLMA-One-Loss-for-More-Accurate-Time-Series-Forecasting.

cs.LG↗

arXiv2025

A Novel Deep Learning Method for Detecting Nucleon-Nucleon Correlations

This study investigates the impact of nucleon-nucleon correlations on heavy-ion collisions using the hadronic transport model SMASH in $\sqrt{s_{\rm NN}}=3$ GeV $^{197}{\rm Au}$+$^{197}{\rm Au}$ collisions. We developed an innovative Monte Carlo sampling method that incorporates both single-nucleon distributions and nucleon-nucleon correlations. By comparing three initial nuclear configurations - a standard Woods-Saxon distribution (un-corr), hard-sphere repulsion (step corr), and ab initio nucleon-nucleon correlations (nn-corr)- we revealed minimal differences in traditional observables except for ultra-central collisions. When distinguishing between un-corr and nn-corr configurations, conventional attention-based point cloud networks and multi-event mixing classifiers failed (accuracy ~50%). To resolve this, we developed a novel deep learning architecture integrating multi-event statistics and high-dimensional latent space feature correlations, achieving 60\% overall classification accuracy, which improved to 70\% for central collisions. This method enables the extraction of subtle nuclear structure signals through statistical analysis in high-dimensional latent space, offering a new paradigm for studying initial-state nuclear properties and quark-gluon plasma characteristics in heavy-ion collisions. It overcomes the limitations of traditional single-event analysis in detecting subtle initial-state differences.

nucl-th↗

arXiv2025

A Multi-Modal Deep Learning Framework for Pan-Cancer Prognosis

Prognostic task is of great importance as it closely related to the survival analysis of patients, the optimization of treatment plans and the allocation of resources. The existing prognostic models have shown promising results on specific datasets, but there are limitations in two aspects. On the one hand, they merely explore certain types of modal data, such as patient histopathology WSI and gene expression analysis. On the other hand, they adopt the per-cancer-per-model paradigm, which means the trained models can only predict the prognostic effect of a single type of cancer, resulting in weak generalization ability. In this paper, a deep-learning based model, named UMPSNet, is proposed. Specifically, to comprehensively understand the condition of patients, in addition to constructing encoders for histopathology images and genomic expression profiles respectively, UMPSNet further integrates four types of important meta data (demographic information, cancer type information, treatment protocols, and diagnosis results) into text templates, and then introduces a text encoder to extract textual features. In addition, the optimal transport OT-based attention mechanism is utilized to align and fuse features of different modalities. Furthermore, a guided soft mixture of experts (GMoE) mechanism is introduced to effectively address the issue of distribution differences among multiple cancer datasets. By incorporating the multi-modality of patient data and joint training, UMPSNet outperforms all SOTA approaches, and moreover, it demonstrates the effectiveness and generalization ability of the proposed learning paradigm of a single model for multiple cancer types. The code of UMPSNet is available at https://github.com/binging512/UMPSNet.

eess.IV↗

arXiv2025

ICFNet: Integrated Cross-modal Fusion Network for Survival Prediction

Survival prediction is a crucial task in the medical field and is essential for optimizing treatment options and resource allocation. However, current methods often rely on limited data modalities, resulting in suboptimal performance. In this paper, we propose an Integrated Cross-modal Fusion Network (ICFNet) that integrates histopathology whole slide images, genomic expression profiles, patient demographics, and treatment protocols. Specifically, three types of encoders, a residual orthogonal decomposition module and a unification fusion module are employed to merge multi-modal features to enhance prediction accuracy. Additionally, a balanced negative log-likelihood loss function is designed to ensure fair training across different patients. Extensive experiments demonstrate that our ICFNet outperforms state-of-the-art algorithms on five public TCGA datasets, including BLCA, BRCA, GBMLGG, LUAD, and UCEC, and shows its potential to support clinical decision-making and advance precision medicine. The codes are available at: https://github.com/binging512/ICFNet.

eess.IV↗

arXiv2024

MindShot: A Few-Shot Brain Decoding Framework via Transferring Cross-Subject Prior and Distilling Frequency Domain Knowledge

Aiming to reconstruct visual stimuli from brain signals, brain decoding has recently made significant progress using functional magnetic resonance imaging (fMRI). However, it still has challenging issues such as substantial individual differences and high data collection costs. To simplify these problems, most methods adopt the per-subject-per-model paradigm, but this greatly limits their applications. In this paper, we design a few-shot brain decoding setting specifically for potential clinical scenarios and propose a novel two-stage decoding framework named MindShot, comprising a Multi-Subject Pretraining (MSP) stage and Fourier-based cross-subject Knowledge Distillation (FKD) stage. Firstly, a MSP framework based on multi-modal contrastive learning is constructed to mine the cross-subject prior. Secondly, the FKD is presented to decrease inter-individual differences while improving the decoding adaptability to new individuals. Our approach achieves high semantic fidelity in visual reconstruction on the largest dataset and has the potential to reduce scanning time by up to 99%. Remarkably, MindShot achieves a CLIP accuracy of 83.6% using only 1.8% of the fMRI-image pairs, surpassing the 77.4% accuracy of the method trained on the entire NSD dataset. This makes it feasible to train large-scale brain decoding frameworks that require less data, facilitating practical applications. The code is available at https://github.com/JSinBUPT/MindShot.

cs.CV↗

arXiv2024

Ultrafast Broadband Strong-Field Tunnelling in Asymmetric Nanogaps for Time-Resolved Nanoscopy

Femtosecond-fast and nanometre-size pulses of electrons are emerging as unique probes for ultrafast dynamics at the nanoscale. Presently, such pulses are achievable only in highly sophisticated ultrafast electron microscopes or equally complex setups involving few-cycle-pulsed lasers with stable carrier-envelope phase (CEP) and nanotip probes. Here, we show that the generation of femtosecond pulses of nanoscale tunnelling electrons can be achieved in any ultrafast optical laboratory, using any (deep-UV to mid-IR) femtosecond laser in combination with photosensitive asymmetric nanogap (PAN) diodes fabricated via easy-to-scale adhesion lithography. The dominant mechanism producing tunnelling electrons in PANs is strong-field emission, which is easily achievable without CEP locking or external bias voltage. We employ PANs to demonstrate ultrafast nanoscopy of metal-halide perovskite quantum dots immobilised inside a 10-nm Al/Au nanogap and to characterise laser pulses across the entire optical region (266-6700 nm). Short electron pulses in PANs open the way towards scalable on-chip femtosecond electron measurements and novel design approaches for integrated ultrafast sensing nanodevices.

physics.optics↗

arXiv2023

Now and Future of Artificial Intelligence-based Signet Ring Cell Diagnosis: A Survey

Signet ring cells (SRCs), associated with a high propensity for peripheral metastasis and poor prognosis, critically influence surgical decision-making and outcome prediction. However, their detection remains challenging even for experienced pathologists. While artificial intelligence (AI)-based automated SRC diagnosis has gained increasing attention for its potential to enhance diagnostic efficiency and accuracy, existing methodologies lack systematic review. This gap impedes the assessment of disparities between algorithmic capabilities and clinical applicability. This paper presents a comprehensive survey of AI-driven SRC analysis from 2008 through June 2025. We systematically summarize the biological characteristics of SRCs and challenges in their automated identification. Representative algorithms are analyzed and categorized as unimodal or multi-modal approaches. Unimodal algorithms, encompassing image, omics, and text data, are reviewed; image-based ones are further subdivided into classification, detection, segmentation, and foundation model tasks. Multi-modal algorithms integrate two or more data modalities (images, omics, and text). Finally, by evaluating current methodological performance against clinical assistance requirements, we discuss unresolved challenges and future research directions in SRC analysis. This survey aims to assist researchers, particularly those without medical backgrounds, in understanding the landscape of SRC analysis and the prospects for intelligent diagnosis, thereby accelerating the translation of computational algorithms into clinical practice.

eess.IV↗

arXiv2023

Boundary-Refined Prototype Generation: A General End-to-End Paradigm for Semi-Supervised Semantic Segmentation

Semi-supervised semantic segmentation has attracted increasing attention in computer vision, aiming to leverage unlabeled data through latent supervision. To achieve this goal, prototype-based classification has been introduced and achieved lots of success. However, the current approaches isolate prototype generation from the main training framework, presenting a non-end-to-end workflow. Furthermore, most methods directly perform the K-Means clustering on features to generate prototypes, resulting in their proximity to category semantic centers, while overlooking the clear delineation of class boundaries. To address the above problems, we propose a novel end-to-end boundary-refined prototype generation (BRPG) method. Specifically, we perform online clustering on sampled features to incorporate the prototype generation into the whole training framework. In addition, to enhance the classification boundaries, we sample and cluster high- and low-confidence features separately based on confidence estimation, facilitating the generation of prototypes closer to the class boundaries. Moreover, an adaptive prototype optimization strategy is proposed to increase the number of prototypes for categories with scattered feature distributions, which further refines the class boundaries. Extensive experiments demonstrate the remarkable robustness and scalability of our method across diverse datasets, segmentation networks, and semi-supervised frameworks, outperforming the state-of-the-art approaches on three benchmark datasets: PASCAL VOC 2012, Cityscapes and MS COCO. The code is available at https://github.com/djh-dzxw/BRPG.

cs.CV↗

arXiv2023

OpenDriver: An Open-Road Driver State Detection Dataset

Among numerous studies for driver state detection, wearable physiological measurements offer a practical method for real-time monitoring. However, there are few driver physiological datasets in open-road scenarios, and the existing datasets suffer from issues such as poor signal quality, small sample sizes, and short data collection periods. Therefore, in this paper, a large-scale multimodal driving dataset, OpenDriver, for driver state detection is developed. The OpenDriver encompasses a total of 3,278 driving trips, with a signal collection duration spanning approximately 4,600 hours. Two modalities of driving signals are enrolled in OpenDriver: electrocardiogram (ECG) signals and six-axis motion data of the steering wheel from a motion measurement unit (IMU), which were recorded from 81 drivers and their vehicles. Furthermore, three challenging tasks are involved in our work, namely ECG signal quality assessment, individual biometric identification based on ECG signals, and physiological signal analysis in complex driving environments. To facilitate research in these tasks, corresponding benchmarks have also been introduced. First, a noisy augmentation strategy is applied to generate a larger-scale ECG signal dataset with realistic noise simulation for quality assessment. Second, an end-to-end contrastive learning framework is employed for individual biometric identification. Finally, a comprehensive analysis of drivers' HRV features under different driving conditions is conducted. Each benchmark provides evaluation metrics and reference results. The OpenDriver dataset will be publicly available at https://github.com/bdne/OpenDriver.

cs.AI↗

arXiv2022

Move aside pentacene: Diazapentacene doped para-terphenyl as a zero-field room-temperature maser with strong coupling for cavity quantum electrodynamics

Masers, the microwave analogue of lasers, promise to deliver ultra-low noise amplification of microwave signals for use in medical MRI imaging and deep-space communication. Research on masers in modern times was rekindled thanks to the discovery of gain media that were operable at room-temperature, eschewing bulky cryogenics that hindered their use. However, besides the two known materials of pentacene doped in para-terphenyl and negatively-charged nitrogen-vacancy defects in diamond, there has been scarce progress in the search for completely new room-temperature gain media. Here we show the discovery of 6,13-diazapentacene doped in para-terphenyl as a maser gain medium that can operate at room-temperature and without an external magnetic field. A measured maser pulse power of -10 dBm shows it is on par with pentacene-doped para-terphenyl in absolute power, while possessing compelling advantages against its pentacene predecessor in that it has a faster amplification startup time, can be excited with longer wavelength light at 620 nm and enjoys greater chemical stability from added nitrogen groups. Furthermore, we show that the maser bursts allow 6,13-diazapentacene-doped para-terphenyl to reach the strong coupling regime for cavity quantum electrodynamics, where it has a high cooperativity of 182. We study the optical and microwave spin dynamics of 6,13-diazapentacene-doped para-terphenyl in order to evaluate its behavior as a maser gain medium, where it features fast intersystem crossing and an advantageously higher triplet quantum yield. Our results pave the way for the future discovery of other similar maser materials and help point to such materials as promising candidates for the study of cavity quantum electrodynamic effects at room-temperature.

quant-ph↗