Searcharxiv⌕ Search

arXiv subjects

Peng Xu

Publications and source records attributed to Peng Xu.

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

Extensible Fluxonium Architecture Using Tunable Couplers with Low Shunt Capacitance

Fluxonium qubits have demonstrated high-fidelity operations and long coherence times in small-scale systems, highlighting their promise for quantum computing. However, large-scale integration into a high-performance two-dimensional (2D) qubit array remains the central challenge for practical applications. In this work, we introduce an extensible architecture for scaling up fluxonium qubits in 2D grids. To address the key challenges, namely achieving controllable strong interaction and high connectivity for qubits featuring small shunting capacitors (footprints), we propose using low-shunt-capacitance couplers to enable tunable interactions between fluxonium qubits. When embedded into 2D square lattices, large couplings can be achieved even with relatively small coupling capacitances, thus enabling multiple connections with sufficient capacitance budget. We further propose coupler realizations based on generalized flux qubit circuits, specifically the quarton and the fluxonium, and demonstrate that both enable fast, high-fidelity gates with low spectator errors, while supporting multiple connections on 2D grids.

quant-ph↗
arXiv2026

Fast entangling gates on fluxoniums via parametric modulation of plasmon interaction

In superconducting quantum processors, exploring diverse control methods could offer essential versatility and redundancy to mitigate challenges such as frequency crowding, spurious couplings, control crosstalk, and fabrication variability, thus leading to better system-level performance. Here we introduce a control strategy for fast entangling gates in a scalable fluxonium architecture, utilizing parametric modulation of the plasmon interaction. In this architecture, fluxoniums are coupled via a tunable coupler, whose transition frequency is flux-modulated to control the inter-fluxonium plasmon interaction. A bSWAP-type interaction is activated by parametrically driving the coupler at the sum frequency of the plasmon transitions of the two fluxoniums, resulting in the simultaneous excitation or de-excitation of both plasmon modes. This strategy therefore allow the transitions between computational states and non-computational plasmon states, enabling the accumulation of conditional phases on the computational subspace and facilitating the realization of controlled-phase gates. By focusing on a specific case of these bSWAP-type interactions, we show that a simple drive pulse enables sub-100ns CZ gates with an error below $10^{-4}$. Given its operational flexibility and extensibility, this approach could potentially offer a foundational framework for developing scalable fluxonium-based quantum processors.

quant-ph↗
arXiv2026

Precision measurements at the interface between unitary and non-unitary encoding

We investigate precision scaling at the interface between unitary and non-unitary encoding under generalized noise including single-particle and collective dephasing and decay. Using linear response theory and the error propagation formula, we derive analytic precision expressions for both the unitary parameter $Ω$ and the dissipation strength $γ$. For unitary encoding, when the observable commutes with a Hermitian noise operator, the optimal encoding time is independent of $N$, yielding the Heisenberg limit $ΔΩ\propto 1 / N$; otherwise the precision degrades to the standard quantum limit or ceases to improve with $N$. For non-unitary encoding, when $[\hat{A}, \hat{O}] = 0$, the precision is insensitive to intrinsic dynamics and encoding time, scaling as $Δγ\propto \sqrt{γ/ \expval*{\hat{L}^\dagger \hat{L}}}$. Notably, for collective decay, the Dicke state reaches the Heisenberg limit $Δγ\propto 1 / N$, demonstrating that entanglement can enhance non-unitary estimation. Our results provide a unified framework and practical guidance for designing quantum metrology protocols in noisy environments.

quant-ph↗
arXiv2026

Rec-Distill: An Industrial Distillation Pipeline for Large-Scale Recommendation Models

Large recommendation models have demonstrated substantial potential gains under scaling laws, yet these gains are difficult to realize in industrial recommendation systems because real-world deployment requires lightweight models with strict serving efficiency and latency guarantees. This creates a fundamental gap between offline model scaling and online deployment. In this work, we present Rec-Distill, an industrial distillation pipeline that transfers the performance gains of large-scale recommendation modeling to efficient serving models. Rec-Distill combines large-teacher scaling with student-side transfer optimization through decoupled training, black-box distillation, debiasing mechanism, and a hybrid batch-streaming pipeline for dynamic recommendation environments. Across multiple recommendation and advertising scenarios on real-world platforms, our framework scales teacher models up to 24B dense parameters and 20K behavior sequence length, while enabling lightweight students to recover a substantial portion of teacher gains, with distillation transferability exceeding 60% in the best setting. Extensive offline and online experiments further show that these transferred gains consistently translate into measurable business improvements under industrial constraints. These results demonstrate that Rec-Distill provides a practical framework for distilling large-scale recommendation models into deployable, cost-efficient serving systems, while also establishing a reliable path toward scaling recommendation models to even larger regimes in the future.

cs.IR↗
arXiv2026

Measuring the Hubble constant with strongly lensed gravitational waves from space-based detector networks

The measurement of the Hubble constant $H_0$ plays a central role in modern cosmology. In this work, we investigate the potential of strongly lensed gravitational-wave (SLGW) signals from massive binary black hole mergers to constrain $H_0$ using future space-based detector networks. We consider two observational scenarios: one in which the source redshift is unknown, and another in which it is independently determined through electromagnetic observations. We show that meaningful constraints on $H_0$ can still be achieved without source-redshift information, provided that the lens redshift is known. For individual SLGW events, the joint Taiji+LISA analysis improves the measurement precision of $H_0$ by approximately a factor of two compared with the Taiji-only configuration. Extending the analysis to the population level, we combine five simulated SLGW events and find that the uncertainty in $H_0$, quantified by the 95\% credible interval, reaches the $1.1\times10^{-1}$ level when the source redshift is treated as unknown, and further improves to $4.2\times10^{-2}$ when the source redshift is independently measured. Our results demonstrate that joint space-based gravitational-wave observations can substantially enhance the cosmological capability of SLGW events and provide a promising avenue for precision measurements of the Hubble constant.

astro-ph.CO↗
arXiv2026

Ringdown Signatures of Dehnen Dark Matter Halos: Fluid Modes and Detectability with Space-Based Detectors

In this work, we investigate the feasibility of using ringdown waveforms from supermassive black holes immersed in dark-matter halos to extract both the intrinsic black-hole parameters and those characterizing the surrounding matter distribution with future space-based gravitational-wave detectors. Building on the fully relativistic framework developed by Cardoso {\it et al.}, in which the dark-matter degrees of freedom are explicitly accounted for by minimal coupling to the gravitational sector, we construct numerical waveforms for a variety of Dehnen-type dark-matter profiles. We then convert these simulated waveforms into realistic data streams for future space-based gravitational-wave observatories, consistently implementing the second-generation Time-Delay Interferometry scheme in the analysis. We calculate the signal-to-noise ratios and perform a Bayesian parameter estimation to infer the model parameters, quantifying their measurability through the resulting posterior distributions. Our results indicate that the presence of dark matter can induce sizable modifications to the waveforms through the appearance of fluid modes at late times. Furthermore, dark-matter profiles with more pronounced spikes leave stronger imprints on the gravitational-wave signal, thereby enhancing the prospects for parameter inference with future space-based detectors such as LISA, Taiji, and TianQin.

gr-qc↗
arXiv2026

LENS: Multi-level Evaluation of Multimodal Reasoning with Large Language Models

Multimodal Large Language Models (MLLMs) have achieved significant advances in integrating visual and linguistic information, yet their ability to reason about complex and real-world scenarios remains limited. The existing benchmarks are usually constructed in the task-oriented manner without guarantee that different task samples come from the same data distribution, thus they often fall short in evaluating the synergistic effects of lower-level perceptual capabilities on higher-order reasoning. To lift this limitation, we contribute Lens, a multi-level benchmark with 3.4K contemporary images and 60K+ human-authored questions covering eight tasks and 12 daily scenarios, forming three progressive task tiers, i.e., perception, understanding, and reasoning. One feature is that each image is equipped with rich annotations for all tasks. Thus, this dataset intrinsically supports to evaluate MLLMs to handle image-invariable prompts, from basic perception to compositional reasoning. In addition, our images are manully collected from the social media, in which 53% were published later than Jan. 2025. We evaluate 15+ frontier MLLMs such as Qwen2.5-VL-72B, InternVL3-78B, GPT-4o and two reasoning models QVQ-72B-preview and Kimi-VL. These models are released later than Dec. 2024, and none of them achieve an accuracy greater than 60% in the reasoning tasks. Project page: https://github.com/Lens4MLLMs/lens. ICCV 2025 workshop page: https://lens4mllms.github.io/mars2-workshop-iccv2025/

cs.CV↗
arXiv2026

Ada-MK: Adaptive MegaKernel Optimization via Automated DAG-based Search for LLM Inference

When large language models (LLMs) serve real-time inference in commercial online advertising systems, end-to-end latency must be strictly bounded to the millisecond range. Yet every token generated during the decode phase triggers thousands of kernel launches, and kernel launch overhead alone can account for 14.6% of end-to-end inference time. MegaKernel eliminates launch overhead and inter-operator HBM round-trips by fusing multiple operators into a single persistent kernel. However, existing MegaKernel implementations face a fundamental tension between portability and efficiency on resource-constrained GPUs such as NVIDIA Ada: hand-tuned solutions are tightly coupled to specific architectures and lack portability, while auto-compiled approaches introduce runtime dynamic scheduling whose branch penalties are unacceptable in latency-critical settings. We observe that under a fixed deployment configuration, the optimal execution path of a MegaKernel is uniquely determined, and runtime dynamic decision-making can be entirely hoisted to compile time. Building on this insight, we propose Ada-MK: (1) a three-dimensional shared-memory constraint model combined with K-dimension splitting that reduces peak shared memory usage by 50%; (2) MLIR-based fine-grained DAG offline search that solidifies the optimal execution path, completely eliminating runtime branching; and (3) a heterogeneous hybrid inference engine that embeds MegaKernel as a plugin into TensorRT-LLM, combining high-throughput Prefill with low-latency Decode. On an NVIDIA L20, Ada-MK improves single-batch throughput by up to 23.6% over vanilla TensorRT-LLM and 50.2% over vLLM, achieving positive gains across all tested scenarios--the first industrial deployment of MegaKernel in a commercial online advertising system.

cs.CL↗
arXiv2026

Efficient LLM-based Advertising via Model Compression and Parallel Verification

Large language models (LLMs) have shown remarkable potential in advertising scenarios such as ad creative generation and targeted advertising. However, deploying LLMs in real-time advertising systems poses significant challenges due to their high inference latency and computational cost. In this paper, we propose an Efficient Generative Targeting framework that integrates adaptive group quantization, layer-adaptive hierarchical sparsification, and prefix-tree parallel verification to accelerate LLM inference while preserving generation quality. Extensive experiments on two real-world advertising scenarios demonstrate that our framework achieves significant speedup with acceptable quality degradation, making it operationally viable for practical deployments.

cs.CL↗
arXiv2026

A Unified Approach for Computing Wasserstein Barycenters of Discrete and Continuous Measures

Computing the unregularized Wasserstein barycenter for measure-valued data is a challenging optimization task. Recent algorithms have been tailored to either discrete measures as point clouds or continuous measures discretized on regular grids. In this work, we propose a primal mirror descent algorithm for computing the exact Wasserstein barycenter in the Fisher-Rao geometry. Our algorithm is a unified approach that is flexible enough to simultaneously cover discrete and absolutely continuous input measures, with convergence guarantees in both settings. In particular, when all input measures are discrete, our algorithm, initialized from any probability density, solves a sequence of semi-discrete optimal transport subproblems and produces absolutely continuous iterates that converge to the discrete barycenter. We use synthetic and real data examples to demonstrate the promising result in terms of accuracy and computational cost.

math.OC↗
arXiv2026

Detection of Lensed Gravitational Waves in the Millihertz Band Using Frequency-Domain Lensing Feature Extraction Network

The space-based gravitational wave (GW) detectors are expected to observe lensed GW events, offering new opportunities for cosmology and fundamental physics.Across the millihertz band, lensing effects transition from the wave-optics regime at lower frequencies to the geometric-optics approximation at higher frequencies.Although traditional GW identification methods, such as matched filtering, are well established and effective, the intense computational resources required motivate the search for more efficient alternatives to accelerate candidate event screening. To address this bottleneck, we introduce a Dual-Channel Lensing feature extraction eXtended Long Short-Term Memory Network (DCL-xLSTM). Unlike conventional recurrent architectures, DCL-xLSTM uses a matrix-valued memory structure and a memory-mixing mechanism to effectively capture amplitude patterns that span the entire millihertz frequency band. Trained on data generated by Point Mass (PM) and Singular Isothermal Sphere (SIS) models accounting for the transition from wave-optics to geometric-optics, the proposed method achieves an area under the curve (AUC) exceeding 0.99, maintaining a true positive rate (TPR) above $98\%$ at a false positive rate (FPR) below $1\%$.The network is robust against variations in signal-to-noise ratio, lens type, and lens mass, establishing its viability as a high-efficiency tool for future space-based GW detection.

astro-ph.IM↗
arXiv2026

A Comparative Evaluation of AI Agent Security Guardrails

This report presents a comparative evaluation of DKnownAI Guard in AI agent security scenarios, benchmarked against three competing products: AWS Bedrock Guardrails, Azure Content Safety, and Lakera Guard. Using human annotation as the ground truth, we assess each guardrail's ability to detect two categories of risks: threats to the agent itself (e.g., instruction override, indirect injection, tool abuse) and requests intended to elicit harmful content (e.g., hate speech, pornography, violence). Evaluation results demonstrate that DKnownAI Guard achieves the highest recall rate at 96.5\% and ranks first in true negative rate (TNR) at 90.4\%, delivering the best overall performance among all evaluated guardrails.

cs.CR↗
arXiv2026

High-Precision Ground Characterization of Test-Mass Magnetic Properties for the Taiji Gravitational Wave Mission via a Physics-Informed Neural Framework

Taiji is a gravitational wave detection mission in space initiated by the Chinese Academy of Sciences, which will open the millihertz window through a heliocentric triangular constellation of three drag-free spacecraft. Its ultimate sensitivity is determined partly by the residual acceleration noise of the gravitational reference sensors (GRS), within which the coupling between the test-mass and the fluctuating environmental magnetic field constitutes one of the key stray-force contributions. Following the path established by the LISA and TianQin teams, high-precision ground characterization of remanent magnetic moment $\vec{m}_r$ and volume susceptibility $χ$ of the test masses is a central step in the Taiji pre-launch test program. A persistent challenge for this characterization is the non-stationary, colored background noise inherent to torsion-pendulum facilities, which systematically biases classical Ordinary Least Squares (OLS) and Kalman filter (KF) estimators. We propose an AI-enhanced Differentiable Weighted Least Squares (AI-WLS) framework that fuses a dilated one-dimensional residual network, acting as a dynamic noise evaluator, with a fully differentiable analytical physical solver. This architecture preserves the exact linear mapping from the magnetic parameters to the torque response while autonomously identifying and suppressing contaminated data segments. Validated on real measured noise from the Changchun Institute of Optics, Fine Mechanics and Physics torsion-pendulum facility developed for Taiji, which achieves a torque sensitivity of order $10^{-13}\,\mathrm{N\cdot m\,Hz^{-1/2}}$, the AI-WLS framework bounds the maximum absolute estimation errors at $4.46\times 10^{-10}\,\mathrm{A\cdot m^2}$ for $\vec{m}_r$ and $7.8\times 10^{-8}$ for $χ$, satisfying Taiji's ground-test requirements on all these parameters simultaneously.

astro-ph.IM↗
arXiv2026

Physics informed operator learning of parameter dependent spectra

Spectral problems governed by differential operators underpin a wide range of physical systems, yet remain computationally challenging because their spectra depend sensitively on continuous parameters and often demand repeated evaluations across parameter space. Here we present $\texttt{DeepOPiraKAN}$, an open source physics informed neural network architecture for spectral analysis. By combining operator learning with enhanced optimization stability, it captures the underlying parameter-to-spectrum mapping in a single model, avoiding repeated spectral solutions at isolated points in parameter space. As a representative and stringent benchmark, we apply this framework to the computation of quasinormal modes of Kerr black holes. A single trained network accurately resolves modes with $(\ell,m)\in \{(2,0),(2,1)\}$ and overtones up to $n=7$ across the full spin range, achieving relative errors of $\mathcal{O}(10^{-6})$ for the fundamental mode and gradually increasing to $\mathcal{O}(10^{-4})$ for higher overtones, benchmarked against the Leaver's method. This level of accuracy is already significant for black hole spectroscopy and practical ringdown modelling for current and future observatories. More broadly, these results highlight the potential of $\texttt{DeepOPiraKAN}$ as a general and scalable framework for parameter dependent spectral problems across complex physical systems.

gr-qc↗
arXiv2026

Coherent Rydberg excitation of single atoms using a pulsed fiber amplifier

In recent years, the growing scale of programmable neutral-atom arrays has led to an increasing demand for higher-power Rydberg excitation light. Although pulsed amplifiers deliver higher peak power than continuous-wave lasers, their use for efficient coherent Rydberg excitation of single atoms in arrays has been limited by challenges such as pulse distortion, synchronization with excitation sequences, and spectral linewidth broadening. Here, we address these issues using a fiber-based master-oscillator power-amplifier system. We demonstrate efficient coherent Rydberg excitation of single atoms in a rubidium atom array, achieving performance comparable to continuous-wave methods. This study provides a potentially new technical pathway toward future large-scale quantum simulation and computation with Rydberg atom arrays.

quant-ph↗
arXiv2026

Bayesian Analysis of Gravitational Wave Microlensing Effects from Galactic Double White Dwarfs

Gravitational waves (GWs) from the galactic double white dwarf (DWD) systems are one of the primary targets for upcoming space-based detectors. Due to their vast abundance and widespread distribution throughout the Galactic disk and bulge, these systems may provide a high-statistical population for probing GW microlensing effects induced by Galactic compact objects. To evaluate the detectability of such effects, in this work we simulate the four-year observation of DWD systems by Taiji, in the form of a second-generation Time Delay Interferometry (TDI) data stream. Within a Bayesian inference framework, we estimate parameters for lensed GWs from DWD systems for different values of the lens parameters, including the lens mass $M_\mathrm{L}\in [10, 10^6]$\,M$_\odot$, the effective velocity $v_\mathrm{eff}\in [50, 500]$\,km/s and the initial separation $L\in [R_\mathrm{E}, 3R_\mathrm{E}]$, and obtain the uncertainties of the corresponding parameters. These results characterize the capability of future Taiji observations to probe such systems. We further employ the Bayesian model selection framework to distinguish between lensed and unlensed scenarios, and investigate the impacts of three key physical parameters of the lens system: $M_\mathrm{L}$, $v_\mathrm{eff}$, and $L$ on distinguishing lensing events. Our results show that when $M_\mathrm{L}$ is below $10^5$\,M$_\odot$ or $L\geq3R_\mathrm{E}$, it is not possible to distinguish between lensed and unlensed models. For $v_\mathrm{eff}$, although the Bayes factor decreases as $v_\mathrm{eff}$ decreases, the lensed and unlensed models can still be distinguished within our parameter range.

astro-ph.GA↗
arXiv2026

Prospect on constraining environment-dependent dilaton model from gravitational redshift measurements

Scalar-tensor gravity represents a natural extension of general relativity. This paper investigates a conformal scalar-tensor gravity, the environmentally dependent dilaton model, and identifies regions of its parameter space potentially constrained by future experiments using atomic clocks to measure gravitational redshift. We propose an experimental scheme in which atomic clocks are placed in environments of different mass densities, such as ultrahigh vacuum, water, or osmium, and their frequency shifts are compared to probe the scalar field contribution to gravitational redshift. By further modeling the mass distribution in low-density environments with a discrete representation, we go beyond the standard continuous approximation. Despite limitations inherent to specific experimental configurations, our analysis reveals that a significant portion of the parameter space remains accessible. Importantly, the accessible regions are complementary to those constrained by existing tests, as they are primarily sensitive to relatively weak couplings. Consequently, high-precision gravitational redshift experiments hold the potential to exclude significant regions of this parameter space in the future.

gr-qc↗
arXiv2026

Long-range tunable coupler for modular fluxonium quantum processors

The path toward practical superconducting quantum processors requires the integration of a large number of high-performance qubits. Modular architectures could offer a way to address the scaling limitations of monolithic designs by partitioning a large quantum processor into physically separated modules, or chiplets, linked through long-range interconnects. In this context, although fluxonium qubits have emerged as a compelling platform for quantum computing due to their long coherence times and high-fidelity gates, existing coupling schemes remain restricted to qubits in close proximity on a single chip. This limitation inherently precludes the long-range interconnects essential for modular integration. In this work, we propose a long-range tunable coupler designed to interconnect fluxonium qubits separated by more than one centimeter, thereby supporting the realization of modular fluxonium quantum processors. Under realistic assumptions, the proposed coupler has the potential to achieve inter-module two-qubit gate performance, specifically sub-100-ns gates with intrinsic errors below $10^{-4}$, comparable to that of intra-module (intra-chiplet) gates, while enabling modular integration with low quantum crosstalk, a key requirement for scalable systems. We further discuss the integration of this coupler into modular fluxonium lattices and explore its feasibility for achieving the higher connectivity and longer coupling range required for complex quantum error correction codes. This work could contribute to the development of large-scale fluxonium quantum processors by bridging their demonstrated potential with modular scalability.

quant-ph↗