arXiv subjects

Ming Wang

Publications and source records attributed to Ming Wang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing

Large language models (LLMs) have achieved impressive performance across natural language tasks and are increasingly deployed in real-world applications. Despite extensive safety alignment efforts, recent studies show that such alignment is often shallow and remains vulnerable to jailbreak attacks. Existing defense mechanisms, including decoding-based constraints and post-hoc content detectors, struggle against sophisticated jailbreaks, often intervening robust detection or excessively degrading model utility. In this work, we examine the decoding process of LLMs and make a key observation: even when successfully jailbroken, models internally exhibit latent safety-related signals during generation. However, these signals are overridden by the model's drive for fluent continuation, preventing timely self-correction or refusal. Building on this observation, we propose a simple yet effective approach that explicitly surfaces and leverages these latent safety signals for early detection of unsafe content during decoding. Experiments across diverse jailbreak attacks demonstrate that our approach significantly enhances safety, while maintaining low over-refusal rates on benign inputs and preserving response quality. Our results suggest that activating intrinsic safety-awareness during decoding offers a promising and complementary direction for defending against jailbreak attacks. Code is available at: https://github.com/zyz13590/SafeProbing.

cs.AI↗

arXiv2026

Symmetry conditions for spacetime observability of wave equations on the torus

We study observability for the one-dimensional wave equation on the torus from spacetime measurable observation sets. While the Geometric Control Condition (GCC) provides a sufficient criterion in many classical settings, it is no longer sufficient in this framework. We construct explicit counterexamples showing the failure of observability despite the validity of GCC. This leads to the introduction of an additional symmetry condition on the observation set, referred to as the Observable Symmetry Condition (OSC). We prove that observability holds if and only if both GCC and OSC are satisfied. We also show that unique continuation holds if and only if both OSC and a weak form of GCC are satisfied.

math.AP↗

arXiv2026

Towards Privacy-Preserving Mental Health Support with Large Language Models

Large language models (LLMs) have shown promise for mental health support, yet training such models is constrained by the scarcity and sensitivity of real counseling dialogues. In this article, we present MindChat, a privacy-preserving LLM for mental health support, together with MindCorpus, a synthetic multi-turn counseling dataset constructed via a multi-agent role-playing framework. To synthesize high-quality counseling data, the developed dialogue-construction framework employs a dual closed-loop feedback design to integrate psychological expertise and counseling techniques through role-playing: (i) turn-level critique-and-revision to improve coherence and counseling appropriateness within a session, and (ii) session-level strategy refinement to progressively enrich counselor behaviors across sessions. To mitigate privacy risks under decentralized data ownership, we fine-tune the base model using federated learning with parameter-efficient LoRA adapters and incorporate differentially private optimization to reduce membership and memorization risks. Experiments on synthetic-data quality assessment and counseling capability evaluation show that MindCorpus improves training effectiveness and that MindChat is competitive with existing general and counseling-oriented LLM baselines under both automatic LLM-judge and human evaluation protocols, while exhibiting reduced privacy leakage under membership inference attacks.

cs.AI↗

arXiv2026

Electric field switching of altermagnetic spin-splitting in multiferroic skyrmions

Magnetic skyrmions are localized magnetic structures that retain their shape and stability over time, thanks to their topological nature. Recent theoretical and experimental progress has laid the groundwork for understanding magnetic skyrmions characterized by negligible net magnetization and ultrafast dynamics. Notably, skyrmions emerging in materials with altermagnetism, a novel magnetic phase featuring lifted Kramers degeneracy-have remained unreported until now. In this study, we demonstrate that BiFeO3, a multiferroic renowned for its strong coupling between ferroelectricity and magnetism, can transit from a spin cycloid to a Neel-type skyrmion under antidamping spin-orbit torque at room temperature. Strikingly, the altermagnetic spin splitting within BiFeO3 skyrmion can be reversed through the application of an electric field, revealed via the Circular photogalvanic effect. This quasiparticle, which possesses a neutral topological charge, holds substantial promise for diverse applications-most notably, enabling the development of unconventional computing systems with low power consumption and magnetoelectric controllability.

cond-mat.mtrl-sci↗

arXiv2026

Fine-grained Verbal Attack Detection via a Hierarchical Divide-and-Conquer Framework

In the digital era, effective identification and analysis of verbal attacks are essential for maintaining online civility and ensuring social security. However, existing research is limited by insufficient modeling of conversational structure and contextual dependency, particularly in Chinese social media where implicit attacks are prevalent. Current attack detection studies often emphasize general semantic understanding while overlooking user response relationships, hindering the identification of implicit and context-dependent attacks. To address these challenges, we present the novel "Hierarchical Attack Comment Detection" dataset and propose a divide-and-conquer, fine-grained framework for verbal attack recognition based on spatiotemporal information. The proposed dataset explicitly encodes hierarchical reply structures and chronological order, capturing complex interaction patterns in multi-turn discussions. Building on this dataset, the framework decomposes attack detection into hierarchical subtasks, where specialized lightweight models handle explicit detection, implicit intent inference, and target identification under constrained context. Extensive experiments on the proposed dataset and benchmark intention detection datasets show that smaller models using our framework significantly outperform larger monolithic models relying on parameter scaling, demonstrating the effectiveness of structured task decomposition.

cs.CL↗

arXiv2026

Warming-driven rise in soil moisture entropy signals destabilization of the Asian Water Tower

The Tibetan Plateau (TP), known as the "Asian Water Tower," is currently undergoing a rapid wetting trend. While this moisture increase is often viewed as beneficial for water availability, it remains unclear whether the hydrological system itself is becoming more resilient or drifting toward instability. Here, we apply an entropy-based framework to quantify the changing structural organization of the TP's soil moisture system. We show that from 2000 to 2024, regional wetting has driven a long-term decline in entropy, reflecting an increase in system order and stability due to enhanced hydrological buffering capacity. This stability is modulated by the El Niño-Southern Oscillation (ENSO), which regulates regional heterogeneity via a distinct spatial dipole. Crucially, however, CMIP6 climate projections reveal an alarming reversal: future warming triggers a rise in entropy. This transition signals a loss of systemic resilience, characterized by intensified spatial disorder and potential abrupt regime shifts by the mid-century. Our findings suggest that while current wetting provides a stabilizing buffer, continued warming is projected to amplify spatial heterogeneity, thereby destabilizing the Asian Water Tower, with significant risks for downstream water security.

physics.ao-ph↗

arXiv2025

A New Integrative Learning Framework for Integrating Multiple Secondary Outcomes into Primary Outcome Analysis: A Case Study on Liver Health

In the era of big data, secondary outcomes have become increasingly important alongside primary outcomes. These secondary outcomes, which can be derived from traditional endpoints in clinical trials, compound measures, or risk prediction scores, hold the potential to enhance the analysis of primary outcomes. Our method is motivated by the challenge of utilizing multiple secondary outcomes, such as blood biochemistry markers and urine assays, to improve the analysis of the primary outcome related to liver health. Current integration methods often fall short, as they impose strong model assumptions or require prior knowledge to construct over-identified working functions. This paper addresses these statistical challenges and potentially opens a new avenue in data integration by introducing a novel integrative learning framework that is applicable in a general setting. The proposed framework allows for the robust, data-driven integration of information from multiple secondary outcomes, promotes the development of efficient learning algorithms, and ensures optimal use of available data. Extensive simulation studies demonstrate that the proposed method significantly reduces variance in primary outcome analysis, outperforming existing integration approaches. Additionally, applying this method to UK Biobank (UKB) reveals that cigarette smoking is associated with increased fatty liver measures, with these effects being particularly pronounced in the older adult cohort.

stat.ME↗

arXiv2025

Contrastive Learning-Based Dependency Modeling for Anomaly Detection in Cloud Services

This paper addresses the challenges of complex dependencies and diverse anomaly patterns in cloud service environments by proposing a dependency modeling and anomaly detection method that integrates contrastive learning. The method abstracts service interactions into a dependency graph, extracts temporal and structural features through embedding functions, and employs a graph convolution mechanism to aggregate neighborhood information for context-aware service representations. A contrastive learning framework is then introduced, constructing positive and negative sample pairs to enhance the separability of normal and abnormal patterns in the representation space. Furthermore, a temporal consistency constraint is designed to maintain representation stability across time steps and reduce the impact of short-term fluctuations and noise. The overall optimization combines contrastive loss and temporal consistency loss to ensure stable and reliable detection across multi-dimensional features. Experiments on public datasets systematically evaluate the method from hyperparameter, environmental, and data sensitivity perspectives. Results show that the proposed approach significantly outperforms existing methods on key metrics such as Precision, Recall, F1-Score, and AUC, while maintaining robustness under conditions of sparse labeling, monitoring noise, and traffic fluctuations. This study verifies the effectiveness of integrating dependency modeling with contrastive learning, provides a complete technical solution for cloud service anomaly detection, and demonstrates strong adaptability and stability in complex environments.

cs.LG↗

arXiv2025

Integrating Structure-Aware Attention and Knowledge Graphs in Explainable Recommendation Systems

This paper designs and implements an explainable recommendation model that integrates knowledge graphs with structure-aware attention mechanisms. The model is built on graph neural networks and incorporates a multi-hop neighbor aggregation strategy. By integrating the structural information of knowledge graphs and dynamically assigning importance to different neighbors through an attention mechanism, the model enhances its ability to capture implicit preference relationships. In the proposed method, users and items are embedded into a unified graph structure. Multi-level semantic paths are constructed based on entities and relations in the knowledge graph to extract richer contextual information. During the rating prediction phase, recommendations are generated through the interaction between user and target item representations. The model is optimized using a binary cross-entropy loss function. Experiments conducted on the Amazon Books dataset validate the superior performance of the proposed model across various evaluation metrics. The model also shows good convergence and stability. These results further demonstrate the effectiveness and practicality of structure-aware attention mechanisms in knowledge graph-enhanced recommendation.

cs.IR↗

arXiv2025

Collaborative Evolution of Intelligent Agents in Large-Scale Microservice Systems

This paper proposes an intelligent service optimization method based on a multi-agent collaborative evolution mechanism to address governance challenges in large-scale microservice architectures. These challenges include complex service dependencies, dynamic topology structures, and fluctuating workloads. The method models each service as an agent and introduces graph representation learning to construct a service dependency graph. This enables agents to perceive and embed structural changes within the system. Each agent learns its policy based on a Markov Decision Process. A centralized training and decentralized execution framework is used to integrate local autonomy with global coordination. To enhance overall system performance and adaptability, a game-driven policy optimization mechanism is designed. Through a selection-mutation process, agent strategy distributions are dynamically adjusted. This supports adaptive collaboration and behavioral evolution among services. Under this mechanism, the system can quickly respond and achieve stable policy convergence when facing scenarios such as sudden workload spikes, topology reconfigurations, or resource conflicts. To evaluate the effectiveness of the proposed method, experiments are conducted on a representative microservice simulation platform. Comparative analyses are performed against several advanced approaches, focusing on coordination efficiency, adaptability, and policy convergence performance. Experimental results show that the proposed method outperforms others in several key metrics. It significantly improves governance efficiency and operational stability in large-scale microservice systems. The method demonstrates strong practical value and engineering feasibility.

cs.DC↗

arXiv2025

DIFNet: Decentralized Information Filtering Fusion Neural Network with Unknown Correlation in Sensor Measurement Noises

In recent years, decentralized sensor networks have garnered significant attention in the field of state estimation owing to enhanced robustness, scalability, and fault tolerance. Optimal fusion performance can be achieved under fully connected communication and known noise correlation structures. To mitigate communication overhead, the global state estimation problem is decomposed into local subproblems through structured observation model. This ensures that even when the communication network is not fully connected, each sensor can achieve locally optimal estimates of its observable state components. To address the degradation of fusion accuracy induced by unknown correlations in measurement noise, this paper proposes a data-driven method, termed Decentralized Information Filter Neural Network (DIFNet), to learn unknown noise correlations in data for discrete-time nonlinear state space models with cross-correlated measurement noises. Numerical simulations demonstrate that DIFNet achieves superior fusion performance compared to conventional filtering methods and exhibits robust characteristics in more complex scenarios, such as the presence of time-varying noise. The source code used in our numerical experiment can be found online at https://wisdom-estimation.github.io/DIFNet_Demonstrate/.

eess.SP↗

arXiv2025

Artificial Intelligence-Based Multiscale Temporal Modeling for Anomaly Detection in Cloud Services

This study proposes an anomaly detection method based on the Transformer architecture with integrated multiscale feature perception, aiming to address the limitations of temporal modeling and scale-aware feature representation in cloud service environments. The method first employs an improved Transformer module to perform temporal modeling on high-dimensional monitoring data, using a self-attention mechanism to capture long-range dependencies and contextual semantics. Then, a multiscale feature construction path is introduced to extract temporal features at different granularities through downsampling and parallel encoding. An attention-weighted fusion module is designed to dynamically adjust the contribution of each scale to the final decision, enhancing the model's robustness in anomaly pattern modeling. In the input modeling stage, standardized multidimensional time series are constructed, covering core signals such as CPU utilization, memory usage, and task scheduling states, while positional encoding is used to strengthen the model's temporal awareness. A systematic experimental setup is designed to evaluate performance, including comparative experiments and hyperparameter sensitivity analysis, focusing on the impact of optimizers, learning rates, anomaly ratios, and noise levels. Experimental results show that the proposed method outperforms mainstream baseline models in key metrics, including precision, recall, AUC, and F1-score, and maintains strong stability and detection performance under various perturbation conditions, demonstrating its superior capability in complex cloud environments.

cs.LG↗

arXiv2025

EvidenceBench: A Benchmark for Extracting Evidence from Biomedical Papers

We study the task of automatically finding evidence relevant to hypotheses in biomedical papers. Finding relevant evidence is an important step when researchers investigate scientific hypotheses. We introduce EvidenceBench to measure models performance on this task, which is created by a novel pipeline that consists of hypothesis generation and sentence-by-sentence annotation of biomedical papers for relevant evidence, completely guided by and faithfully following existing human experts judgment. We demonstrate the pipeline's validity and accuracy with multiple sets of human-expert annotations. We evaluated a diverse set of language models and retrieval systems on the benchmark and found that model performances still fall significantly short of the expert level on this task. To show the scalability of our proposed pipeline, we create a larger EvidenceBench-100k with 107,461 fully annotated papers with hypotheses to facilitate model training and development. Both datasets are available at https://github.com/EvidenceBench/EvidenceBench

cs.CL↗

arXiv2025

PTQ1.61: Push the Real Limit of Extremely Low-Bit Post-Training Quantization Methods for Large Language Models

Large Language Models (LLMs) suffer severe performance degradation when facing extremely low-bit (sub 2-bit) quantization. Several existing sub 2-bit post-training quantization (PTQ) methods utilize a mix-precision scheme by leveraging an unstructured fine-grained mask to explicitly distinguish salient weights, while which introduces an extra 1-bit or more per weight. To explore the real limit of PTQ, we propose an extremely low-bit PTQ method called PTQ1.61, which enables weight quantization to 1.61-bit for the first time. Specifically, we first introduce a one-dimensional structured mask with negligibly additional 0.0002-bit per weight based on input activations from the perspective of reducing the upper bound of quantization error to allocate corresponding salient weight channels to 4-bit. For non-salient channels binarization, an efficient block-wise scaling factors optimization framework is then presented to take implicit row-wise correlations and angular biases into account. Different from prior works that concentrate on adjusting quantization methodologies, we further propose a novel paradigm called quantization preprocessing, where we argue that transforming the weight distribution of the pretrained model before quantization can alleviate the difficulty in per-channel extremely low-bit PTQ. Extensive experiments indicate our PTQ1.61 achieves state-of-the-art performance in extremely low-bit quantization. Codes are available at https://github.com/zjq0455/PTQ1.61.

cs.LG↗

arXiv2025

The periodic KdV with control on space-time measurable sets

In this paper, we establish the local exact controllability of the KdV equation on torus around equilibrium states, where both the spatial control region and the temporal control region are sets of positive measure. The proof is based on a novel strategy for proving observability inequalities on space-time measurable sets. This approach is applicable to a broad class of dispersive equations on torus.

math.AP↗

arXiv2025

An Enhanced Privacy-preserving Federated Few-shot Learning Framework for Respiratory Disease Diagnosis

The labor-intensive nature of medical data annotation presents a significant challenge for respiratory disease diagnosis, resulting in a scarcity of high-quality labeled datasets in resource-constrained settings. Moreover, patient privacy concerns complicate the direct sharing of local medical data across institutions, and existing centralized data-driven approaches, which rely on amounts of available data, often compromise data privacy. This study proposes a federated few-shot learning framework with privacy-preserving mechanisms to address the issues of limited labeled data and privacy protection in diagnosing respiratory diseases. In particular, a meta-stochastic gradient descent algorithm is proposed to mitigate the overfitting problem that arises from insufficient data when employing traditional gradient descent methods for neural network training. Furthermore, to ensure data privacy against gradient leakage, differential privacy noise from a standard Gaussian distribution is integrated into the gradients during the training of private models with local data, thereby preventing the reconstruction of medical images. Given the impracticality of centralizing respiratory disease data dispersed across various medical institutions, a weighted average algorithm is employed to aggregate local diagnostic models from different clients, enhancing the adaptability of a model across diverse scenarios. Experimental results show that the proposed method yields compelling results with the implementation of differential privacy, while effectively diagnosing respiratory diseases using data from different structures, categories, and distributions.

cs.LG↗

arXiv2025

Cascaded frequency conversion of highly charged femtosecond spatiotemporal optical vortices

The degrees of freedom inherent in spatiotemporal optical vortices (STOV's) afford intriguing opportunties to manipulate complex light fields for broad applications such as optical communication, light-matter interactions, particle manipulation, quantum optics, and electron acceleration in the relativistic regime. Unlike previous studies examining the second harmonic generation (SHG) of STOV's having an input topological charge (TC) l^{\left(ω\right)}=1, here we experimentally demonstrate cascaded second and third harmonic generation of STOV's to achieve more diverse wavelength selectivity via sum frequency generation (SFG), achieving unprecedented TC values up to l^{\left(ω\right)}=40. The large TC values are attributed to second (third) harmonic generation of an incident beam satisfying l^{\left(2ω\right)}=2l^{\left(ω\right)} (l^{\left(3ω\right)}=3l^{\left(ω\right)}). What is more, the wavelength of the generated STOV field was found to be tunable by controlling the position of the phase singularity in the frequency domain, and optimizing the nonlinear phase matching condition. Our experimental measurements extend the principle of conservation of the spatiotemporal topological charge to general nonlinear optical parametric processes, suggesting a fundamental approach to the production of STOV fields of arbitrarily large TC and at arbitrary visible wavelengths, and beyond (e.g., the ultraviolet range).

physics.optics↗

arXiv2025

Satellite Signal Detection via Rydberg-Atom Receiver

Rydberg-atom receivers aim for ultra-high sensitivity to microwave fields through various techniques, but receiving satellite signals has remained a significant challenge, due to the difficulty of capturing weak microwaves over long distances. In this work, we introduce a high-gain antenna to focus satellite signals, and then apply into an atomic cell via a microwave cavity. Using microwave-enhanced coupling, the minimum detectable power of incident microwave is down to -128 dBm, and the corresponding sensitivity is estimated as 21 nV/cm/Hz1/2 at frequency of 3.80 GHz. Furthermore, beacon signal from geostationary satellites is captured with Rydberg sensors for the first time, without the need for a low-noise amplifier. And C-band modulated signals are read out with a signal-to-noise ratio of 8 dB. Our results mark a significant breakthrough in facilitating satellite communications using Rydberg-atom receivers.

quant-ph↗