arXiv subjects

Yan Guo

Publications and source records attributed to Yan Guo.

At least 19 recordsLinked to original sources

arXiv2026

Determinants and Limits of LLM Security-Tool Orchestration: A Study with HexStrike-AI

Large language model agents driving security tool suites over the Model Context Protocol are increasingly common. Yet the factors that bound their capability remain poorly characterized: how much depends on the model versus the client that drives it, whether constraining the agent to the orchestrator's own tools helps, and where capability is limited by reasoning rather than by missing tools. Using HexStrikeAI, an open-source orchestrator that exposes 150+ tools, as a testbed, we follow a methodology that evaluates the system, diagnoses its failures, and applies targeted improvements. We run 86 picoCTF challenges across seven categories and three difficulty tiers, under three tool-access regimes and three model/client configurations (774 trials). We then apply corrections to existing tools, agent-behavior changes, and eleven new capability tools, and re-run the previously-unsuccessful trials. The diagnosis isolates the driving client as a first-order factor for a fixed model (a 2.1 * gap between two DeepSeek clients) and a monotonic difficulty gradient, with the largest gains in the mid tier. The overall solve rate rises from 55.4% to 72.0%, and every configuration improves significantly (paired McNemar p < 0.001, non-overlapping 95% confidence intervals). The residual failures are reasoning- or environment-bound rather than missing-tool. A 60-run stability sub-study finds single-run verdicts reproducible (17/20 unanimous). We discuss what the results imply for how such orchestrators should be evaluated, and we are explicit about the limits: the study uses a single benchmark, the fixes were tuned on the same challenges they were evaluated on, and the client effect is demonstrated for one model only, so its generality to other models remains a hypothesis.

cs.SE↗

arXiv2026

Decoupling Reconnaissance and Exploitation: Measuring the Capability Boundaries of LLM-Based Web Penetration Testing

Large Language Models (LLMs) have shown promise for automated penetration testing, yet existing end-to-end black-box evaluations are highly susceptible to error cascading: failures in early reconnaissance can mask an agent's actual ability to exploit vulnerabilities. To more accurately characterize these capabilities, we propose a two-stage decoupled evaluation framework that separates exploit execution from reconnaissance. Using ground-truth injection and knowledge-driven ablation across 70 high-fidelity web vulnerability testbeds, our framework isolates exploitation performance from reconnaissance noise. We empirically evaluate five open-source penetration-testing agents, covering multiagent, monolithic, and graph-driven architectures, on a strictly aligned subset of 50 representative vulnerabilities. The results reveal a substantial capability gap. With accurate vulnerability context, agents achieve a functional success rate of up to 90.0%, whereas autonomous reconnaissance, measured by targeted vulnerability recall, plateaus at approximately 50.0%, primarily due to failures in parsing unstructured telemetry. Cross-architectural analysis further reveals distinct capability niches: multi-agent isolation is more effective for long-sequence interactions such as de-serialization, while monolithic and graph-driven designs perform better on short-chain injections and cross-session access-control vulnerabilities, respectively. This decoupled evaluation work provides a fine-grained benchmarking protocol and an empirical basis for designing next-generation automated offensive security agents.

cs.CR↗

arXiv2026

SemGeoNav:A Safety-Guided Visual Navigation Approach with Semantic Reasoning and Geometric Planning

Learning-based visual navigation has enhanced semantic goal-reaching capabilities. However, due to their black-box nature, purely end-to-end models often lack explicit geometric constraints, leading to unpredictable and unreliable obstacle avoidance in open environments. Conversely, traditional geometric planners ensure safety but struggle with high-dimensional visual targets. To address these limitations, we propose SemGeoNav, a novel hierarchical visual navigation framework.It tightly integrates the high-level semantic reasoning of end-to-end models with the reliable local planning ability of geometry-based methods, achieving robust image-based navigation while significantly improving obstacle avoidance. Furthermore, we introduce a temporal trajectory smoothing mechanism to ensure continuous and stable robot motion. We evaluated SemGeoNav on a Unitree Go2 quadruped robot in real-world environments. The results demonstrate that SemGeoNav outperforms existing representative methods, including ViNT and NoMaD, achieving higher success rates and shorter navigation times.

cs.RO↗

arXiv2026

Sharp $H_{x}^{s}$ Ill-posedness of the Hard-sphere Boltzmann Equation

We investigate the ill-posedness mechanism of the hard-sphere Boltzmann equation in $H_{x}^{s}$ Sobolev space. Via a direct construction, we prove a strong-weak type ill-posedness result in the low-regularity regime $s<1$, establishing a sharp threshold in connection to the local $s>1$ well-posedness result [11]. Instead of originating from the large-velocity growth of the collision kernel, this illposedness is generated by the loss term and dispersive effects. Consequently, we prove a dispersion-driven nonlinear instability mechanism for the hard-sphere Boltzmann equation, and provide a capstone of the ill-posedness series [18,20].

math.AP↗

arXiv2026

Argo: Efficient Importance Labeling for Enterprise Email Systems

Email importance labeling has long been a critical yet challenging problem for businesses and individuals. Traditional approaches; such as keyword matching, user-defined rules, and sender-based heuristics; demand extensive manual feature engineering and fail to scale effectively or generalize. Recent advances in large language models (LLMs) demonstrate strong potential and a natural fit for this task, offering deep contextual understanding and superior labeling quality. However, using LLM models like GPT-4.1 at enterprise email volumes incurs prohibitive computational costs and hinders real-world deployment. We explore the trade-off space of using alternative labeling schemes as opposed to GPT4.1 scale LLMs, with the goal of achieving near GPT level labeling quality with significantly lower cost. We develop Argo, an enterprise email labeling framework, where we construct a profiler to efficiently search the cost quality trade-off space of labeling and identify cost-efficient alternatives to labeling emails. Additionally, we design an on-demand provisioning scheme to intelligently scale Argo with real time load, to minimize cost increases during peak load inference. Over 3 open-source email datasets, Argo achieves 148-167X inference cost reduction with negligible quality degradation and 20-640000X lower profiling costs, making large-scale, context-aware email labeling practical for enterprises.

cs.MA↗

arXiv2026

Effective Noise Mitigation via Quantum Circuit Learning in Quantum Simulation of Integrable Spin Chains

We propose a noise-mitigation quantum simulation strategy for near-term quantum devices based on Quantum Circuit Learning (QCL), which is in particular effective for integrable quantum spin chains. The method trains a shallow variational circuit to approximate a deeper time-evolution circuit by learning the conserved charges and only a small amount of dynamical information in the system. Under realistic noise models, the learned circuit maintains both conserved quantities and dynamical observables significantly closer to their true values than the noisy simulation of the original circuit. We demonstrate, on small-scale prototypes, that QCL can act as an effective, physics-informed error mitigation strategy, producing shorter, more robust circuits without exponential sampling overhead.

quant-ph↗

arXiv2026

Stability of Inflow Problem for Hyperbolic Systems

Inflow BC plays a critical role in the study of hyperbolic PDE in a bounded domain. We establish $W^{1,\infty}$ stability for 1D hyperbolic conservation laws with inflow data in a bounded interval, and $W^{2,3+}$ stability of a large class of shear flows for the 3D incompressible Euler system with inflow BC in finite square or circular pipes.

math.AP↗

arXiv2026

TimeRFT: Stimulating Generalizable Time Series Forecasting for TSFMs via Reinforcement Finetuning

Time Series Foundation Models (TSFMs) have demonstrated strong generalization capability and data efficiency in time series forecasting through large-scale pretraining. However, adapting TSFMs to downstream forecasting tasks remains challenging due to temporal distribution shifts and varying data availability. Specifically, the non-stationary and uncertain nature of time series data leads to discrepancies between historical training and future forecasting distributions, making existing Supervised FineTuning (SFT)-based adaptation vulnerable to overfitting and limited generalization. Moreover, forecasting tasks often operate under varying data regimes, requiring TSFMs to extract generalizable temporal patterns from limited training samples. To address these challenges, we propose Time series Reinforcement FineTuning (TimeRFT), a reinforcement learning-based adaptation paradigm for TSFMs. TimeRFT introduces two forecasting-oriented training recipes: (i) A quality-aware temporal reward mechanism providing fine-grained credit assignment by holistically evaluating the contribution of each prediction step to overall forecasting performance. (ii) A difficulty-aware data selection strategy prioritizing informative time series samples with generalizable forecasting patterns. Extensive experiments on diverse real-world forecasting benchmarks demonstrate that TimeRFT consistently surpasses SFT-based adaptation methods across various real-world forecasting tasks with different data regimes, achieving improved prediction accuracy and enhanced generalization against unforeseen distribution shifts.

eess.SP↗

arXiv2026

Asymptotic stability of symmetric flows with viscous inflow boundary condition

We study the two-dimensional incompressible Navier-Stokes equations in a channel $\Omega=(0,L)\times(0,H)$ with small viscosity $\varepsilon\ll1$, an $\varepsilon$-Navier slip condition on the horizontal walls, and a viscous inflow condition for the perturbation stream function. For a broad class of symmetric base profiles $u_0(y)$ vanishing on the walls, we construct an exact steady solution $(u_s,v_s)$ that is $O(\varepsilon^{1/3})$-close to the shear $(u_0,0)$. We then develop a new weighted vorticity energy method to prove uniform linear stability and exponential decay: perturbations decay exponentially in a weighted $L^2$ norm on the time scale $O(\varepsilon^{-1/3})$. In the short-channel regime $L\ll1$, the method yields nonlinear asymptotic stability with threshold $O(\varepsilon^{2/3})$. In the long-channel regime, assuming concavity together with a spectral condition, we introduce a quantity \textit{Rayleigh vorticity} to control the non-favorable terms and obtain nonlinear stability with threshold $O(\varepsilon^{5/6+})$.

math.AP↗

arXiv2026

Accelerating Vision Transformers on Brain Processing Unit

With the advancement of deep learning technologies, specialized neural processing hardware such as Brain Processing Units (BPUs) have emerged as dedicated platforms for CNN acceleration, offering optimized INT8 computation capabilities for convolutional operations. Meanwhile, Vision Transformer (ViT) models, such as the Data-efficient Image Transformer (DeiT), have demonstrated superior performance and play increasingly crucial roles in computer vision tasks. However, due to the architectural mismatch between CNN-optimized hardware and Vision Transformer computation characteristics--namely, that linear layers in Transformers operate on three-dimensional data while BPU acceleration is designed for four-dimensional convolution operations-it is difficult or even impossible to leverage BPU's advantages when deploying Vision Transformers. To address this challenge, we propose a novel approach that restructures the Vision Transformer by replacing linear layers and layer normalization operations with carefully designed convolutional operators. This enables DeiT to fully utilize the acceleration capabilities of BPUs, while allowing the original weight parameters to be inherited by the restructured models without retraining or fine-tuning. To the best of our knowledge, this is the first successful deployment of Vision Transformers that fully leverages BPU classification datasets demonstrate the effectiveness of our approach. Specifically, the quantized DeiT-Base model achieves 80.4% accuracy on ImageNet, compared to the original 81.8%, while obtaining up to a 3.8* inference speedup. Our finetuned DeiT model on the flower classification dataset also achieves excellent performance, with only a 0.5% accuracy drop for the DeiT-Base model, further demonstrating the effectiveness of our method.

cs.CV↗

arXiv2025

Machine learning identification of fractional-order vortex beam diffraction process

Fractional-order vortex beams possess fractional orbital angular momentum (FOAM) modes, which theoretically have the potential to increase transmission capacity infinitely. Therefore, they have significant application prospects in the fields of measurement, optical communication and micro-particle manipulation. However, when fractional-order vortex beams propagate in free space, the discontinuity of the helical phase makes them susceptible to diffraction in practical applications, thereby affecting the accuracy of OAM mode recognition and severely limiting the use of FOAM-based optical communication. Achieving machine learning recognition of fractional-order vortex beams under diffraction conditions is currently an urgent and unreported issue. Based on ResNet, a deep learning (DL) method of accurately recognizing the propagation distance and topological charge of fractional-order vortex beam diffraction process is proposed in this work. Utilizing both experimentally measured and numerically simulated intensity distributions, a dataset of vortex beam diffraction intensity patterns in atmospheric turbulence environments is created. An improved 101-layer ResNet structure based on transfer learning is employed to achieve accurate and efficient recognition of the FOAM model at different propagation distances. Experimental results show that the proposed method can accurately recognize FOAM modes with a propagation distance of 100 cm, a spacing of 5 cm, and a mode spacing of 0.1 under turbulent conditions, with an accuracy of 99.69%. This method considers the effect of atmospheric turbulence during spatial transmission, allowing the recognition scheme to achieve high accuracy even in special environments. It has the ability to distinguish ultra-fine FOAM modes and propagation distances, which cannot be achieved by traditional methods.

physics.optics↗

arXiv2025

Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learning

Advanced deep learning-based approaches have been actively applied to forecast the spatiotemporal physical dynamics governed by partial differential equations (PDEs), which acts as a critical procedure in tackling many science and engineering problems. As real-world physical environments like PDE system parameters are always capricious, how to generalize across unseen out-of-distribution (OOD) forecasting scenarios using limited training data is of great importance. To bridge this barrier, existing methods focus on discovering domain-generalizable representations across various PDE dynamics trajectories. However, their zero-shot OOD generalization capability remains deficient, since extra test-time samples for domain-specific adaptation are still required. This is because the fundamental physical invariance in PDE dynamical systems are yet to be investigated or integrated. To this end, we first explicitly define a two-fold PDE invariance principle, which points out that ingredient operators and their composition relationships remain invariant across different domains and PDE system evolution. Next, to capture this two-fold PDE invariance, we propose a physics-guided invariant learning method termed iMOOE, featuring an Invariance-aligned Mixture Of Operator Expert architecture and a frequency-enriched invariant learning objective. Extensive experiments across simulated benchmarks and real-world applications validate iMOOE's superior in-distribution performance and zero-shot generalization capabilities on diverse OOD forecasting scenarios.

cs.LG↗

arXiv2025

Steady Prandtl Layer for Compressible Fluids

Despite its importance, there have been few PDE results to investigate Prandtl layers for compressible fluids, in which the thermal boundary layer for the temperature field interacts with the classical velocity Prandtl boundary layer in a strong nonlinear fashion. We establish local-in-$x$ well-posedness of Prandtl layers for steady compressible fluids and construct Prandtl layer expansion in strong Sobolev norms, a foundation for the validity of Prandtl layer theory \cite{Guo-Wang}. Our method is based on a combination of the recent quotient estimate for the velocity field and an introduction of a pseudo entropy estimate for the temperature field.

math.AP↗

arXiv2025

Nonlinear stability of the Larson-Penston collapse

We prove nonlinear stability of the Larson-Penston family of self-similarly collapsing solutions to the isothermal Euler-Poisson system. Our result applies to radially symmetric perturbations and it is the first full nonlinear stability result for radially imploding compressible flows. At the heart of the proof is the ground state character of the Larson-Penston solution, which exhibits important global monotonicity properties used throughout the proof. One of the key challenges is the proof of mode-stability for the non self-adjoint spectral problem which arises when linearising the dynamics around the Larson-Penston collapsing solution. To exclude the presence of complex growing modes other than the trivial one associated with time translation symmetry, we use a high-order energy method in low and high frequency regimes, for which the monotonicity properties are crucially exploited, and use rigorous computer-assisted techniques in the intermediate regime. In addition, the maximal dissipativity of the linearised operator is proven on arbitrary large backward light cones emanating from the singular point using the global monotonicity of the Larson-Penston solutions. Such a flexibility in linear analysis also facilitates nonlinear analysis and allows us to identify the exact number of derivatives necessary for the nonlinear stability statement. The proof is based on a two-tier high-order weighted energy method which ties bounds derived from the Duhamel formula to quasilinear top order estimates. To prove global existence we further use the Brouwer fixed point theorem to identify the final collapse time, which suppresses the trivial instability caused by the time-translation symmetry of the system.

math.AP↗

arXiv2025

Validity of Prandtl Expansion for Steady Compressible Navier-Stokes-Fourier Flows

Assume no-slip boundary conditions for the velocity field and either insulated or Dirichlet boundary conditions for the temperature field in a steady compressible fluid. In the inviscid limit $\v \rightarrow 0$, we develop a mathematical framework for the uniform-in-$\v$ remainder estimate for the linear steady compressible Navier-Stokes-Fourier equations around a Prandtl layer profile with both velocity and thermal layers, which leads to the validity of the Prandtl layer expansion.

math.AP↗

arXiv2024

Diffusive Expansion of the Boltzmann equation for the flow past an obstacle

The exterior domain problem is essential in fluid and kinetic equations. In this paper, we establish the validity of the diffusive expansion for the Boltzmann equations to the Navier-Stokes-Fourier system up to the critical time in an exterior domain with non-zero passing flow. We apply the $L^3-L^6$ framework to the unbounded domain in this paper.

math.AP↗

arXiv2024

Small scales in inviscid limits of steady fluids

In this article, we study the 2D incompressible steady Navier-Stokes equation in a channel $(-L,0)\times(-1,1)$ with the no-slip boundary condition on $\{Y = \pm 1\}$, and consider the inviscid limit $\varepsilon \to 0$. In the special case of Euler shear flow $(u_e(Y),0)$, we construct a steady Navier-Stokes solution for $\varepsilon \ll1$, $$\left\{ \begin{aligned} &u^\varepsilon \sim u_e + u_p + O(\sqrt{\varepsilon}),\\ &v^\varepsilon \sim h(Y) \exp\{Xu_e(Y)/\varepsilon\} + O(\sqrt{\varepsilon}), \end{aligned}\right. $$ where $u_p$ represents the classical Prandtl layer profile, and $h(Y)$ is an arbitrary smooth, compactly-supported function with small magnitude. While the classical Prandtl boundary layer $u_p$ exhibits a small scale of order $\sqrt{\varepsilon}$ in $Y$ near $Y = \pm 1$, the profile we construct reveals an $\varepsilon$ small scale of $Xu_e(Y)$ in the vertical velocity component.

math.AP↗

arXiv2024

EdgeNAT: Transformer for Efficient Edge Detection

Transformers, renowned for their powerful feature extraction capabilities, have played an increasingly prominent role in various vision tasks. Especially, recent advancements present transformer with hierarchical structures such as Dilated Neighborhood Attention Transformer (DiNAT), demonstrating outstanding ability to efficiently capture both global and local features. However, transformers' application in edge detection has not been fully exploited. In this paper, we propose EdgeNAT, a one-stage transformer-based edge detector with DiNAT as the encoder, capable of extracting object boundaries and meaningful edges both accurately and efficiently. On the one hand, EdgeNAT captures global contextual information and detailed local cues with DiNAT, on the other hand, it enhances feature representation with a novel SCAF-MLA decoder by utilizing both inter-spatial and inter-channel relationships of feature maps. Extensive experiments on multiple datasets show that our method achieves state-of-the-art performance on both RGB and depth images. Notably, on the widely used BSDS500 dataset, our L model achieves impressive performances, with ODS F-measure and OIS F-measure of 86.0%, 87.6% for multi-scale input,and 84.9%, and 86.3% for single-scale input, surpassing the current state-of-the-art EDTER by 1.2%, 1.1%, 1.7%, and 1.6%, respectively. Moreover, as for throughput, our approach runs at 20.87 FPS on RTX 4090 GPU with single-scale input. The code for our method will be released soon.

cs.CV↗