Searcharxiv⌕ Search

arXiv subjects

Ning Wang

Publications and source records attributed to Ning Wang.

At least 55 records · Page 3Linked to original sources
arXiv2026

Ising Dirac fermions across a topological phase transition

Dirac fermions have attracted significant interest due to their relativistic dispersions and close connections to topological physics, yet they are generally expected to be gapped in two-dimensional systems with strong Ising spin orbit coupling, making their realization in such materials an outstanding challenge. Here we report the emergence of six fold degenerate Dirac fermions in an Ising moire system across a quantum spin Hall transition in twisted WSe2. In a 3.65 degree device, we observe a quantum spin Hall phase at high electric fields with nearly quantized resistance h/(2e2), and a Dirac semimetal phase over a broad range of electric fields near zero field. Magnetotransport measurements of the Dirac phase exhibit a half-integer Landau fan sequence, characteristic of Dirac fermions, with six-fold degeneracy on the hole-doped side and two fold degeneracy on the electron-doped side. Temperature dependence shows weakly metallic behavior consistent with a semimetallic state. Our twist-angle-dependent transport measurements map out a complete phase diagram and identify a critical twist angle of 3.3 degree, establishing the phase boundary between the quantum spin Hall and Dirac semimetal regimes. Our work establishes a new route to realizing Dirac fermions in strongly spin orbit coupled moire systems through a topological phase transition, providing a promising platform for high mobility spintronics.

cond-mat.mes-hall↗
arXiv2026

Observation of a Mott quantum spin Hall insulator in twisted WSe2

Quantum spin Hall (QSH) insulators and Mott insulators are conventionally regarded as distinct insulating phases, arising from band topology and strong Coulomb interactions, respectively. Here, we report the observation of QSH edge transport in a magnetic-field-stabilized Mott insulating state at half filling of the second moire band in a 2.29 degree twisted WSe2 device. This state exhibits a resistance plateau identical to that of the single-particle QSH state at full filling of the first moire valence band, indicating the same number of helical edge channels. Electrical transport measurements reveal nearly quantized resistance that is insensitive to vertical electric field, out-of-plane magnetic field, and temperature below 5 K. Pronounced nonlocal transport and strong negative in-plane magnetoconductance further support helical edge conduction, establishing robust edge transport in the strongly correlated regime. Temperature-dependent Hall measurements reveal a characteristic temperature scale of approximately 10 K, corresponding to an energy scale of about 1 meV. Our results demonstrate that spin-conserved QSH edge states can persist in a half filled, strongly correlated insulating phase and under external magnetic field, opening a route toward interaction-resilient topological transport in moire quantum materials.

cond-mat.mes-hall↗
arXiv2026

Correlation enhanced resistance hysteresis near half filling in MoS2/WSe2 heterobilayer

Ferroelectricity, typically arising from ionic displacements in noncentrosymmetric lattices, enabling applications in memory devices and sensors. Recent advances in two-dimensional materials and van der Waals heterostructures have revealed novel ferroelectric phenomena, including sliding ferroelectricity and correlation-driven ferroelectricity in moire superlattices. In this work, we fabricate and study a MoS2/WSe2 moire superlattice device exhibiting a high field-effect mobility of 17,650 $cm^2V^{-1}s^{-1}$. Electrical transport measurements reveal correlated insulating states accompanied by a prominent and reproducible resistance hysteresis near half filling. Temperature and displacement field dependence further confirms the correlation-enhanced nature of the hysteresis. Our analysis suggests that displacement field-induced metal-to-insulator transition at correlated insulating state coupled with interfacial dipoles enables the observed resistance hysteresis. These results establish correlation enhanced resistance hysteresis near half filling in a MoS2/WSe2 heterobilayer, offering opportunities for exploring emergent quantum phases and device functionalities.

cond-mat.mes-hall↗
arXiv2026

Biased Random Walk on $\mathbb Z_+$ with Traps of Linearly Increasing Depth

We study a $λ$-biased random walk $(X_n)_{n\ge0}$ on the deterministic infinite rooted tree $\mathcal{T}=\{(i,j): i\ge0,\,0\le j\le i\}$, whose backbone is $\{(i,0):i\ge0\}$ and, for each $i\ge1$, the segment $\{(i,j):1\le j\le i\}$ forms a trap attached to $(i,0)$. The trapping effect induces long sojourns, yielding asymptotics markedly different from simple random walks. The walk is recurrent for $λ\ge1$ and transient for $0<λ<1$. In the transient regime it is sub-ballistic: its distance from the root grows logarithmically, with \[ \liminf_{n\to\infty}\frac{|X_n|}{\log n}=\frac{1}{\log(1/λ)},\quad \limsup_{n\to\infty}\frac{|X_n|}{\log n}=\frac{2}{\log(1/λ)},\quad\text{a.s.}. \] A contrast between spatial and temporal regeneration emerges. Let $C(n)$ be the number of cutpoints among the first $n$ backbone vertices and $M(N)$ the number of cut times up to time $N$. Then \[ \lim_{n\to\infty}\frac{C(n)}{n}= 1-λ,\qquad \lim_{N\to\infty}\frac{M(N)}{\log N}=\frac{1-λ}{\log(1/λ)},\quad\text{a.s.}, \] so cutpoints have positive linear density while cut times grow only logarithmically.

math.PR↗
arXiv2026

A superconducting surface-code processor with lattice-surgery logical operations

Fault-tolerant logical operations are fundamental for scalable quantum computation. Here, we report the experimental realization of lattice-surgery operations between a pair of distance-three surface-code logical qubits on a planar superconducting processor. During repeated syndrome extraction cycles, the logical qubits exhibit per-cycle error rates of $0.0365(2)$ and $0.0282(1)$, respectively, after leakage events are rejected. By leveraging joint initialization and lattice splitting, we deterministically prepare a logical Bell state, confirming genuine bipartite entanglement via the error-corrected logical state fidelity. We further execute a two-qubit Deutsch-Jozsa algorithm at the logical level to demonstrate algorithmic utility in a fault-tolerant framework. Finally, to achieve universal control, we implement magic-state injection and gate teleportation to realize continuous non-Clifford rotations about the logical $X$ axis. For the logical $R_{X}(π/4)$ gate, we achieve a logical gate fidelity of $0.943_{-9}^{+10}$ conditioned on the absence of detected errors. These results establish lattice surgery as a practical and versatile paradigm for logical computation in near-term surface-code architectures, representing a critical milestone toward scalable fault-tolerant quantum advantage in superconducting circuits.

quant-ph↗
arXiv2026

Practical advantage of non-Hermitian enhanced quantum sensing

Non-Hermitian systems have emerged as a powerful paradigm for ultrasensitive sensing, leveraging unique spectral and dynamical properties unmatched in Hermitian physics. While recent theoretical bounds suggest these protocols offer no metrological advantage over Hermitian ones in the ideal shot-noise-limited regime when rigorously accounting for the success probability of non-unitary evolution, their practical utility in realistic experimental conditions has not yet been systematically explored. In this work, we shift the focus toward practical laboratory performance and demonstrate that non-Hermitian sensing protocols can significantly outperform their Hermitian counterparts in the presence of pervasive classical technical noises. This performance gain mainly stems from a strongly enhanced susceptibility that amplifies the signal response, effectively overcoming the precision floor imposed by technical imperfections. By numerically evaluating the Fisher information under technical noise, we further substantiate the regimes where non-Hermitian platforms yield definitive practical gains. Our results reconcile the ongoing debate between fundamental limits and experimental observations, offering a concrete avenue for building high-precision, noise-resilient sensors.

quant-ph↗
arXiv2026

Lean Meets Theoretical Computer Science: Scalable Synthesis of Theorem Proving Challenges in Formal-Informal Pairs

Formal theorem proving (FTP) has emerged as a critical foundation for evaluating the reasoning capabilities of large language models, enabling automated verification of mathematical proofs at scale. However, progress has been constrained by limited datasets due to the high cost of manual curation and the scarcity of challenging problems with verified formal-informal correspondences. We propose leveraging theoretical computer science (TCS) as a scalable source of rigorous proof problems, where algorithmic definitions enable automated generation of arbitrarily many challenging theorem-proof pairs. We demonstrate this approach on two TCS domains: Busy Beaver problems, which involve proving bounds on Turing machine halting behavior, and Mixed Boolean Arithmetic problems, which combine logical and arithmetic reasoning. Our framework automatically synthesizes problems with parallel formal (Lean4) and informal (Markdown) specifications, creating a scalable pipeline for generating verified proof challenges. Evaluation on frontier models reveals substantial gaps in automated theorem proving: while DeepSeekProver-V2-671B achieves 57.5\% success on Busy Beaver problems, it manages only 12\% on Mixed Boolean Arithmetic problems. These results highlight the difficulty of long-form proof generation even for problems that are computationally easy to verify, demonstrating the value of TCS domains for advancing automated reasoning research.

cs.LO↗
arXiv2026

Model Forensics in AI-Native Wireless Networks: Taxonomy, Applications, and Case Study

As artificial intelligence (AI) is increasingly embedded in wireless networks, models are becoming core components that influence signal processing, resource scheduling and network control. However, model anomalies, tampering and malicious functions also introduce new security risks. In this article, we focus on model forensics in AI-native wireless networks. Specifically, we first discuss key problems including model authenticity verification, malicious function identification and accountability tracing, and summarize the main categories of model forensics. We then explain the role of model forensics in AI-native wireless networks and review representative application scenarios. In the case study, we use RF fingerprinting as an example and present two concrete workflows based on watermark authentication and backdoor detection, illustrating how provenance authentication and malicious behavior identification can be implemented in practice. The results show that model forensics can provide important support for anomaly assessment, provenance tracing and trustworthy operation in AI-native wireless networks. Finally, we outline several promising directions for future research in this emerging area.

cs.CR↗
arXiv2026

Denoising-GS: Gaussian Splatting with Spatial-aware Denoising

Recent advances in 3D Gaussian Splatting (3DGS) have achieved remarkable success in high-fidelity Novel View Synthesis (NVS), yet the optimization process inevitably introduces noisy Gaussian primitives due to the sparse and incomplete initialization from Structure-from-Motion (SfM) point clouds. Most existing methods focus solely on adjusting the positions of primitives during optimization, while neglecting the underlying spatial structure. To this end, we introduce a new perspective by formulating the optimization of 3DGS as a primitive denoising process and propose Denoising-GS, a spatial-aware denoising framework for Gaussian primitives by taking both the positions and spatial structure into consideration. Specifically, we design an optimizer that preserves the spatial optimization flow of primitives, facilitating coherent and directed denoising rather than random perturbations. Building upon this, the Spatial Gradient-based Denoising strategy jointly considers the spatial supports of primitives to ensure gradient-consistent updates. Furthermore, the Uncertainty-based Denoising module estimates primitive-wise uncertainty to prune redundant or noisy primitives, while the Spatial Coherence Refinement strategy selectively splits primitives in sparse regions to maintain structural completeness. Experiments conducted on three benchmark datasets demonstrate that Denoising-GS consistently enhances NVS fidelity while maintaining representation compactness, achieving state-of-the-art performance across all benchmarks. Source code and models will be made publicly available.

cs.CV↗
arXiv2026

Test of Time: Rethinking Temporal Signal of Benchmark Contamination

Post-cutoff performance decay of LLMs has been widely interpreted as a temporal signal for benchmark contamination, where public information released before the training cutoff may have been included into training corpora and inflated model performance by memorization. We critically examine this view and demonstrate that this temporal signal is highly sensitive to how benchmark questions are constructed, even if the underlying source material remains invariant. Specifically, we show that LLM-transformed questions can produce remarkably different temporal patterns compared to fill-in-the-blank (cloze) questions directly retrieved from the very same documents. We validate this effect on prior benchmarks that report clear post-cutoff decay (LiveCodeBench), and show that a simple LLM-driven transformation of the same problems can effectively remove the temporal pattern. We further provide a mechanistic understanding of this phenomenon using influence function analysis. Overall, our results suggest that post-cutoff performance decay is a sensitive contamination signal, motivating more robust contamination probes for reliable LLM evaluation.

cs.AI↗
arXiv2026

Mixture of Heterogeneous Grouped Experts for Language Modeling

Large Language Models (LLMs) based on Mixture-of-Experts (MoE) are pivotal in industrial applications for their ability to scale performance efficiently. However, standard MoEs enforce uniform expert sizes,creating a rigidity that fails to align computational costs with varying token-level complexity. While heterogeneous expert architectures attempt to address this by diversifying expert sizes, they often suffer from significant system-level challenges, specifically unbalanced GPU utilization and inefficient parameter utilization, which hinder practical deployment. To bridge the gap between theoretical heterogeneity and robust industrial application, we propose Mixture of Heterogeneous Grouped Experts (MoHGE) which introduces a two-level routing mechanism to enable flexible, resource-aware expert combinations. To optimize inference efficiency, we propose a Group-Wise Auxiliary Loss, which dynamically steers tokens to the most parameter-efficient expert groups based on task difficulty. To address the critical deployment challenge of GPU load balancing, we introduce an All-size Group-decoupling Allocation strategy coupled with an Intra-Group Experts Auxiliary Loss. These mechanisms collectively ensure uniform computation distribution across GPUs. Extensive evaluations demonstrate that MoHGE matches the performance of MoE architectures while reducing the total parameters by approximately 20% and maintaining balanced GPU utilization. Our work establishes a scalable paradigm for resource-efficient MoE design, offering a practical solution for optimizing inference costs in real-world scenarios. The code is publicly available at https://github.com/UnicomAI/MoHGE.

cs.CL↗
arXiv2026

GLM-5 Serving Parameter Tuning for OpenClaw: Single-Deployment MaaS Inference Optimization for Long-Context Agent Workloads

OpenClaw requests are dominated by long, tool-augmented prefixes, including system prompts, conversation history, and tool outputs fed back into the context window. For this workload, with about 28k-30k input tokens and 500 output tokens per request, serving quality is governed by throughput, TTFT, and tail latency rather than short-prompt throughput alone. This report studies GLM-5 serving-parameter tuning within a MaaS multi-model inference optimization architecture. The scope is the Single-Node Optimization block of the inference-optimization layer, where chunked prefill, tensor parallelism (TP), pipeline parallelism (PP), and request concurrency are tuned for one GLM-5 serving deployment; in this report, "Single-Node Optimization" denotes the architecture block, while experiments run on a two-node, sixteen-GPU cluster. Within the tested space, the best configuration is chunked-prefill-size=3072, tp=4, pp-size=4, and max-running-requests=24. Compared with the conservative 2048/4/4/16 baseline, it increases request throughput from 0.43 to 0.48 req/s and total token throughput from 9029.64 to 9993.23 tok/s, while reducing average TTFT from 8.98 to 6.69 s and latency P90 from 40.23 to 32.64 s. Under the same hardware footprint, this corresponds to an estimated 10.4% lower serving cost per request and 9.6% lower cost per token. The results show that the optimum is workload-specific: larger chunk sizes and deeper queueing do not monotonically improve performance. We therefore recommend 3072 / tp4 / pp4 / max24 as the default OpenClaw deployment profile.

cs.DC↗
arXiv2026

ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying

Large Language Model (LLM) agents have achieved rapid adoption and demonstrated remarkable capabilities across a wide range of applications. To improve reasoning and task execution, modern LLM agents would incorporate memory modules or retrieval-augmented generation (RAG) mechanisms, enabling them to further leverage prior interactions or external knowledge. However, such a design also introduces a group of critical privacy vulnerabilities: sensitive information stored in memory can be leaked through query-based attacks. Although feasible, existing attacks often achieve only limited performance, with low attack success rates (ASR). In this paper, we propose ADAM, a novel privacy attack that features data distribution estimation of a victim agent's memory and employs an entropy-guided query strategy for maximizing privacy leakage. Extensive experiments demonstrate that our attack substantially outperforms state-of-the-art ones, achieving up to 100% ASRs. These results thus underscore the urgent need for robust privacy-preserving methods for current LLM agents.

cs.CR↗
arXiv2026

ALDEN: Boosting Private Data Extraction from Retrieval-Augmented Generation Systems via Active Learning and Distribution Estimation

Retrieval-Augmented Generation (RAG) is widely used to augment large language models with external knowledge retrieval to improve reliability and generalization. However, recent studies have shown that RAG systems remain vulnerable to data extraction attacks, where adversaries can extract private data by embedding malicious commands into user queries. Despite their feasibility, existing attacks typically suffer from low data extraction rates and limited practical effectiveness. Here, we propose ALDEN, a novel attack that effectively and efficiently extracts private data from RAGs. First, we employ active learning to diversify malicious queries and improve data extraction rates. Second, we observe that the data distribution of the underlying knowledge base provides valuable guidance for query generation and introduce a decay-based dynamic algorithm to estimate the corresponding topic distribution. By combining them together, we demonstrate that ALDEN substantially outperforms state-of-the-art methods through comprehensive evaluations.

cs.IR↗
arXiv2026

Beamforming Feedback as a Novel Attack Surface for Wi-Fi Physical-Layer Security

With the rapid evolution of wireless technologies, Wi-Fi has expanded beyond its original role in data transmission to support various emerging applications, particularly in physical-layer security, including device authentication, user authentication, and secret key generation. Despite extensive research on Wi-Fi Channel State Information (CSI)-based physical-layer security, its vulnerabilities remain largely unexplored. In this work, we propose BFIAttack, a novel attack that exploits Beamforming Feedback Information (BFI) to reconstruct the CSI of a legitimate user or device, thereby compromising Wi-Fi-based physical-layer security. We realize the attack by leveraging a closed-form CSI reconstruction method for the single-antenna station scenario and a maximum likelihood estimation-based CSI reconstruction for the multi-antenna station scenario. Moreover, we exploit spatial similarities among antenna pairs to refine the reconstructed CSI and enhance attack effectiveness. Experimental results show that BFIAttack achieves an average attack success rate of $73\%$ in multi-antenna station scenarios with no more than five attack attempts, and over $93\%$ in single-antenna station scenarios with only a single attempt. BFIAttack reveals critical vulnerabilities in existing Wi-Fi-based physical-layer security.

cs.CR↗
arXiv2026

Quantum exciton solid with embedded electron-hole solids in double-layer WSe2

We studied double-layer WSe2 stacked on opposite sides of thin layers of hexagonal Boron nitride with different densities of electrons and holes. For a fixed hole density, the Coulomb drag resistance is found to exhibit plateaus approximately equal to $-h/(4e^2)$ and $-h/(2e^2)$ as the electron density is changed. When the number of electrons is equal to the number of holes, an exciton solid forms whose transport of quantum edge defects gives rise to the drag resistance. When the electron and hole densities are different, the excess electrons form a solid embedded in the exciton solid. The Coulomb drag resistance of the exciton solid comes from the one-dimensional transport of the two lowest energy channels of quantum edge vacancy-interstitial pairs. This corresponds to the first plateau. With the embedded solid, one of these channels is blocked. This corresponds to the second plateau. Transport experiments in the Corbino geometry with no edges and extra heavier holes were carried out. The plateaus disappeared. Three peaks in the resistance at different hole densities were observed. We interpret that the three peaks correspond to the commensurate exciton and two classes of hole solids. We performed phonon calculations of these states and found that the stability of these exciton-based quantum solids shows good agreement with experiment. Our results establish classes of extreme quantum solid states, opening additional avenues for the study of strongly correlated quantum transport phenomena involving quantum defect states.

cond-mat.mes-hall↗
arXiv2026

SkeletonContext: Skeleton-side Context Prompt Learning for Zero-Shot Skeleton-based Action Recognition

Zero-shot skeleton-based action recognition aims to recognize unseen actions by transferring knowledge from seen categories through semantic descriptions. Most existing methods typically align skeleton features with textual embeddings within a shared latent space. However, the absence of contextual cues, such as objects involved in the action, introduces an inherent gap between skeleton and semantic representations, making it difficult to distinguish visually similar actions. To address this, we propose SkeletonContext, a prompt-based framework that enriches skeletal motion representations with language-driven contextual semantics. Specifically, we introduce a Cross-Modal Context Prompt Module, which leverages a pretrained language model to reconstruct masked contextual prompts under guidance derived from LLMs. This design effectively transfers linguistic context to the skeleton encoder for instance-level semantic grounding and improved cross-modal alignment. In addition, a Key-Part Decoupling Module is incorporated to decouple motion-relevant joint features, ensuring robust action understanding even in the absence of explicit object interactions. Extensive experiments on multiple benchmarks demonstrate that SkeletonContext achieves state-of-the-art performance under both conventional and generalized zero-shot settings, validating its effectiveness in reasoning about context and distinguishing fine-grained, visually similar actions.

cs.CV↗
arXiv2026

SparVAR: Exploring Sparsity in Visual AutoRegressive Modeling for Training-Free Acceleration

Visual AutoRegressive (VAR) modeling has garnered significant attention for its innovative next-scale prediction paradigm. However, mainstream VAR paradigms attend to all tokens across historical scales at each autoregressive step. As the next scale resolution grows, the computational complexity of attention increases quartically with resolution, causing substantial latency. Prior accelerations often skip high-resolution scales, which speeds up inference but discards high-frequency details and harms image quality. To address these problems, we present \textbf{SparVAR}, a training-free acceleration framework that exploits three properties of VAR attention: \textbf{(i) strong attention sinks}, \textbf{(ii) cross-scale activation similarity}, and \textbf{(iii) pronounced locality}. Specifically, we dynamically predict the sparse attention pattern of later high-resolution scales from a sparse decision scale, and construct scale self-similar sparse attention via an efficient index-mapping mechanism, enabling high-efficiency sparse attention computation at large scales. Furthermore, we propose cross-scale local sparse attention and implement an efficient block-wise sparse kernel, which achieves $\mathbf{> 5\times}$ faster forward speed than FlashAttention. Extensive experiments demonstrate that the proposed SparVAR can reduce the generation time of an 8B model producing $1024\times1024$ high-resolution images to the \textbf{1s}, \textbf{without skipping the last scales}. Compared with the VAR baseline accelerated by FlashAttention, our method achieves a $\mathbf{1.57\times}$ speed-up while preserving almost all high-frequency details. When combined with existing scale-skipping strategies, SparVAR attains up to a $\mathbf{2.28\times}$ acceleration, while maintaining competitive visual generation quality. Code is available at \href{https://github.com/CAS-CLab/SparVAR}{SparVAR}.

cs.CV↗