arXiv subjects

Ying Zhang

Publications and source records attributed to Ying Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Air-Ground Collaborative Robots for Fire and Rescue Missions: A Survey from the Mapping and Navigation Perspective

Air-ground collaborative robots have shown great potential in the field of fire and rescue. Mapping and navigation, as the key foundation for air-ground collaborative robots to achieve efficient task execution, have attracted a great deal of attention. This growing interest in collaborative robot mapping and navigation is conducive to enhancing the intelligent execution of fire and rescue tasks, but there has been no comprehensive investigation of this field to highlight their strengths. In this paper, we present a systematic review of the air-ground collaborative robots for fire and rescue from a new perspective of mapping and navigation. First, an air-ground collaborative robots framework for fire and rescue missions based on unmanned aerial vehicle (UAV) mapping and unmanned ground vehicle (UGV) navigation is introduced. Then, the research progress of mapping and navigation under this framework is systematically summarized, including UAV mapping, UAV/UGV co-localization, and UGV navigation, with their main achievements and limitations. Based on the needs of fire and rescue missions, the collaborative robots with different numbers of UAVs and UGVs are classified, and their practicality in fire and rescue tasks is elaborated, with a focus on the discussion of their merits and demerits. In addition, the application examples of air-ground collaborative robots in various firefighting and rescue scenarios are given. Finally, this paper emphasizes the current challenges and potential research opportunities, providing a reference for practitioners and researchers interested in this rapidly evolving field of air-ground collaborative robotics.

cs.RO↗

arXiv2026

Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures

Detectors of alignment failures screen deployed language models and score alignment benchmarks. Most are generative judges that spend a decoding pass on every criterion, and classifiers that read token probabilities, such as Llama Guard, still score one fixed label per call. Jev, a model trained with reinforcement learning for calibrated decisions (RLCD), answers many typed questions about one input with calibrated probabilities in a single call. Whether it detects alignment failures has not been measured. We present RLCDAlignBench, which benchmarks Jev on ten alignment failures: sycophancy, jailbreaks, deception, prompt injection, hallucination, privacy violation, social bias, reward hacking, concealing uncertainty, and power seeking. It spans 44 benchmarks and five target models, labelled by each benchmark's scorer and, on two, by humans. Many of these failures are relational, defined against a reference, such as the user's belief or an injected instruction, that the response alone does not reveal. Our key idea is therefore to vary what Jev is asked separately from what it sees: the question's wording and answer type on one side, the fields of the input on the other. A single generic question reaches a median AUROC of 0.886 zero-shot and beats supervised baselines on most benchmarks. Question wording matters little, while context matters more, mostly through fields that encode the label. Jev matches the reference scorer's agreement with human labels, surfaces label defects in existing benchmarks, and costs 63x less than LLM-judge scorers. Code and data: https://github.com/sumleo/RLCDAlignBench.

cs.AI↗

arXiv2026

Exploring the Parameter Space of pvCD-Bonn Potentials under Constraints from Nuclear Saturation Properties

We explore the parameter space of the pvCD-Bonn B potential within the relativistic Brueckner--Hartree--Fock framework under the constraints from the empirical saturation properties of symmetric nuclear matter. We first examine the effects of the scalar-meson and pion coupling constants. Changing the $σ$-meson couplings in the $^{1}S_0$ and $^{3}S_1$--$^{3}D_1$ channels alone cannot reproduce the empirical saturation density and binding energy at the same time, whereas increasing $g_π$ moves the saturation point toward the empirical region. However, changing $g_π$ alone also affects the deuteron properties and tensor-sensitive observables. We therefore extend the parameter search by varying the $ρ$-meson tensor coupling $f_ρ/g_ρ$ and readjusting the effective $σ$-meson couplings in the $^{1}S_0$ and $^{3}P_0$ channels. Four representative parameter sets with $g_π^2/4π=14.7$--15.0 give deuteron properties close to the experimental values and maintain a reasonable description of the main neutron--proton phase shifts and differential cross sections at $E_{\rm lab}=50$ and 212 MeV. Their saturation densities lie in the range $0.159$--$0.161~\mathrm{fm}^{-3}$, with saturation energies between $-15.00$ and $-15.38~\mathrm{MeV}$. For the interactions obtained by varying $g_π$, the neutron-star mass--radius relations show only a weak dependence on the pion coupling, with maximum masses of about $2.24$--$2.28\,M_\odot$ and radii of about $12.3$--$12.6~\mathrm{km}$ at $1.4\,M_\odot$. These results show that including the $ρ$-meson tensor coupling and partial-wave-dependent $σ$-meson couplings provides a better balance between free-space two-nucleon observables and nuclear-matter saturation properties than varying the pion coupling alone.

nucl-th↗

arXiv2026

ResTest: Targeted Coverage of Residual Not-Covered Code Using Large Language Models for Web GUI Testing

Automated web GUI testing (AWGT) approaches explore web applications through GUI actions to achieve code coverage. However, existing approaches, whether random-based, model-based, or reinforcement-learning-based, often struggle to generate continuous and semantically meaningful action sequences for testing complex functionalities, limiting their achievable code coverage. Recent LLM-based approaches partially alleviate this problem but still fall short due to limited capability in inferring testable functionalities and low success rates in executing tasks on complex web applications. In this paper, we propose ResTest, a complementary approach that uses coverage-report-guided LLMs to target residual not-covered code left by existing AWGT approaches. ResTest first runs an existing AWGT approach to broadly explore the application while constructing a state transition graph with summarized information. Once coverage plateaus, ResTest employs an LLM to infer not-covered functionalities based on the state transition graph and coverage report, and then utilizes a specialized LLM-driven GUI agent to execute these functionalities in a targeted manner. Our evaluation on ten open-source web applications shows that ResTest improves three categories of AWGT approaches by 17.52\% to 24.36\% in average code coverage. Ablation studies further confirm the sustained coverage improvement capability and the contribution of each component.

cs.SE↗

arXiv2026

Periodic Tilings of Cardinality Twice a Prime or Nine in Arbitrary Dimension

We prove that every finite translational tile of $\mathbb Z^d$ of cardinality $2q$, where $q$ is an odd prime, or of cardinality nine admits a fully periodic tiling complement. The result holds in every dimension, regardless of the rank of the subgroup generated by the differences of points of the tile. The proofs combine coprime dilation, vanishing sums of roots of unity, and periodic two-colorings. After translating the tile, both arguments restrict to its intrinsic lattice, and every periodic complement constructed there extends to the ambient lattice. For cardinality $2q$, spectral filtering in the intrinsic lattice yields either a lattice complement or a periodic twofold covering compatible with a tiling complement. This compatibility allows the covering to be split by a periodic proper two-coloring. For nine points, intrinsic ranks $1$ and $2$ are treated separately; in higher intrinsic rank, the algebraic reduction in the intrinsic lattice yields either a lattice complement or spectral support on finitely many rational affine circles. These results also give a decision algorithm for tileability in both cardinality families. The circle-supported alternative is resolved by a conditional-density dichotomy and a periodic replacement argument for the exceptional half-density components.

math.CO↗

arXiv2026

HYDRA: Proactive Android Malware Drift Adaptation via Hierarchical Graph Contrastive Learning

Concept drift, driven by the rapid evolution of Android malware, severely degrades the performance of machine learning detectors. Current adaptation strategies are often reactive, responding only after performance has dropped and imposing a significant manual annotation burden, or they are proactive but rely on unstable adversarial training and incomplete, single-level graph representations. To overcome these limitations, we propose HYDRA (Hybrid Drift Adaptation), a proactive adaptation framework that learns drift-invariant representations from hierarchically structured data. HYDRA first models applications using a hybrid graph structure, combining fine-grained Control Flow Graphs (CFGs) and coarse-grained Function Call Graphs (FCGs) to capture comprehensive behavioral patterns. It then introduces a novel cross-domain contrastive learning objective that aligns historical (source) and new (target) data distributions. By generating pseudo-labels for unlabeled target samples, our method pulls representations of semantically similar applications together, regardless of their domain, within a single, stable optimization process. This approach unifies feature learning and domain alignment, eliminating the need for complex adversarial objectives. Extensive experiments on large-scale, time-ordered malware datasets demonstrate that HYDRA achieves substantially lower False Negative and False Positive Rates than state-of-the-art baselines while requiring up to 87.5% fewer labeled samples. Our work thus offers a robust and efficient solution to combat concept drift in security applications.

cs.CR↗

arXiv2026

MedGPT-oss: Training a General-Purpose Vision-Language Model for Biomedicine

Biomedical multimodal assistants have the potential to unify radiology, pathology, and clinical-text reasoning, yet a critical deployment gap remains: top-performing systems are either closed-source or computationally prohibitive, precluding the on-premises deployment required for patient privacy and PHI compliance. We introduce MEDGPT-OSS, an open-weight, 20B-parameter generalist vision-language model designed to facilitate open research in clinical AI. Rather than relying on architectural complexity, MEDGPT-OSS pairs the GPT-oss language backbone with a visual front-end via a optimized, three-stage training curriculum. By progressively domain-adapting these modules through rigorous data curation and long-context multimodal alignment, we demonstrate that a 20B model can bridge the capacity gap. It successfully outperforms larger open medical models on out-of-distribution (OOD) multimodal reasoning and complex text-only clinical tasks. By unifying diverse modalities under a single instruction-following interface, MEDGPT-OSS maintains a parameter-efficient footprint fully compatible with commodity GPUs. We release the complete training recipe, open-weight checkpoints, and a rigorous evaluation harness to serve as a verifiable foundation for privacy-preserving, institution-specific clinical AI research.

cs.CL↗

arXiv2026

RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation

Platform abuse campaigns conceal redirection instructions with emojis, homophones, character decomposition, and redundant symbols, then route users through disguised links to services associated with pornography, fraud, gambling, or illicit transactions. Existing benchmarks evaluate obfuscated text and risky webpages separately, obscuring how target recovery affects downstream evidence acquisition. We introduce RiskChainBench, pairing 3,600 synthetic token-text restoration inputs from 600 source sessions with 600 corresponding human-labeled local web environments. A model first restores the message, operational intent, and destination; the same underlying model then acts as a VLM-driven web agent that investigates the correctly associated website and produces a frozen, evidence-cited risk report without message-side semantics or domain-reputation cues. We score restoration and correct-routing web investigation separately and compose them offline by applying the frozen primary-entry prediction as a gate to the same Task 2 result. Human labels determine task correctness, while a fixed multimodal evidence judge assesses faithfulness, sufficiency, completeness, and consistency. Across ten models, Entry Top-1 ranges from 35.2% to 95.2% and web decision accuracy from 26.3% to 62.8%; the leading systems differ across entry recovery, full reconstruction, website decisions, and fine-grained typing. Execution failures account for 31.9% of web runs, whereas post-decision type errors account for only 0.9%, identifying stable exploration and risk judgment as the principal bottlenecks. We release the benchmark, protocol, and resettable local sandbox.

cs.CL↗

arXiv2026

Ambiguity Function Analysis of OFDM Signals With Pilots and Data Payloads

Practical orthogonal frequency division multiplexing (OFDM) communication frames contain both deterministic pilots and random data payloads, motivating the joint ambiguity function (AF) analysis of the two components when the entire frame is reused for integrated sensing and communication (ISAC). This paper characterizes two discrete AF formulations for different Doppler regimes, namely the discrete periodic AF (DP-AF) and fast-slow-time AF (FST-AF), and derives closed-form expressions for their expected squared values. For the FST-AF, the expected sidelobe level (ESL) is uniform over the delay-Doppler plane and depends only on the pilot count, constellation kurtosis and total number of time-frequency resources, but not on the pilot symbols or pattern. For the DP-AF, we establish attainable lower and upper ESL bounds and show that no pilot design can minimize all sidelobes simultaneously. We further prove that attaining the lower bound at non-zero Doppler requires a periodic pilot pattern, while equally spaced chirp pilots, including Zadoff-Chu (ZC) sequences, maximize the numbers of sidelobes attaining the lower and upper bounds simultaneously. Two representative ZC pilot patterns widely encountered in communication frames are then examined: contiguous placement produces delay-Doppler ridges described by squared Dirichlet kernels, whereas equally spaced placement generates periodic peak-and-notch structures. Both regular patterns exhibit pronounced high sidelobes, suggesting that communication-oriented pilot patterns should be re-designed for delay-Doppler estimation in the context of ISAC. Numerical results validate the analysis and show that irregular pilot placement can suppress high sidelobes and improve target estimation performance.

eess.SP↗

arXiv2026

Enhancing Large Language Model-Based Systems for End-to-End Circuit Analysis Problem Solving

LLMs have shown strong performance in data-rich domains such as programming, but their reliability in engineering tasks remains limited. Circuit analysis is particularly challenging because it requires both multimodal understanding and precise mathematical reasoning. This paper presents an enhanced end-to-end circuit problem-solving framework using Gemini 2.5 Pro as the backbone model for scalable engineering-education applications. We systematically evaluate Gemini 2.5 Pro on undergraduate circuit-analysis problems and identify two major failure modes: circuit-recognition hallucinations, especially source-polarity errors, and reasoning-process hallucinations, such as incorrect current-direction assumptions. To reduce recognition errors, we integrate a fine-tuned YOLO detector with OpenCV-based processing to isolate voltage and current sources for polarity re-identification. To mitigate reasoning errors, we introduce an ngspice-driven verification loop that supports iterative refinement with optional human feedback. On 83 problems, the proposed pipeline achieves 97.59% accuracy, compared with 79.52% for baseline Gemini. Across four hand-drawn diagram variations, accuracy improves from 60.61%--71.21% to 89.39%--92.42%, with statistically significant gains (p<0.005). On 43 problems from a different textbook, accuracy increases from 58.14% to 83.72%, further supporting cross-textbook generalizability. Error analysis shows that circuit recognition remains the dominant source of residual failures, particularly under varying diagram representations. Overall, the framework substantially improves the robustness, scalability, and generalizability of LLM-based circuit problem solving for engineering education and practical circuit analysis.

cs.CY↗

arXiv2026

A Unified Vision-Language Model for PSMA PET/CT Report Generation, Visual Question Answering, and Lesion Segmentation

Accurate PSMA PET/CT interpretation is central to prostate cancer management, yet existing PET/CT AI models typically address isolated tasks. We propose a unified PSMA PET/CT vision-language model for report generation, visual question answering, and lesion segmentation. The framework adopts an LLaVA-style architecture, comprising a PET/CT vision encoder, an MLP-Mixer projection module, a LoRA-tuned large language model, and a 3D segmentation branch. Training followed a four-stage strategy: vision encoder pretraining, projection-layer alignment, VLM fine-tuning, and final multitask tuning. Language tasks used 5,747 PSMA PET/CT datasets with paired reports, while segmentation used the PSMA subset of AutoPET. The model outperformed PET2REP and a CT-based baseline across standard report-generation metrics, improved performance across VQA question types, and achieved higher Dice and lesion-level overlap F1 than SegAnyPET and nnUNet. These results support the feasibility of a unified framework for structured, interactive, interpretable PSMA PET/CT analysis with voxel-level grounding within a single multitask model architecture.

cs.CV↗

arXiv2026

Fuglede's Conjecture for Cyclic Groups of Square-free Order

We prove Fuglede's conjecture for finite cyclic groups of square-free order. Specifically, if $N$ is square-free and $A\subseteq \mathbb Z_N$ is nonempty with $|A|=m$, then $A$ is spectral if and only if it tiles $\mathbb Z_N$; moreover, these conditions hold precisely when $m\mid N$ and $A$ is a complete set of representatives modulo $m$. This gives a complete structural description of spectral and tiling sets in $\mathbb Z_N$. The key new ingredient is a prime-coordinate elimination theorem: if $P\nmid nm$ is an odd prime, then every $m$-point spectral pair in $\mathbb Z_n\times\mathbb Z_P$ projects injectively to a spectral pair in $\mathbb Z_n$. Its proof combines character-matrix orthogonality, cyclotomic coefficient comparison, and a $P$-adic divisibility argument for associated Gram matrices. Iterating this reduction yields the result for arbitrary square-free $N$, with no restriction on the number or relative sizes of its prime factors.

math.CO↗

arXiv2026

Periodic Joint Co-tiles of Independent Tiles

For every $d\geq2$, we prove that $d-1$ independent finite tiles in $\Z^d$ admit a fully periodic joint co-tile whenever they admit a joint co-tile. Here independence means that the tiles contain the origin and every choice of one nonzero vector from each tile is linearly independent. This removes property $(\star)$ from the periodic existence theorem of Meyerovitch, Sanadhya and Solomon and recovers Bhattacharya's theorem when $d=2$. More generally, we prove periodic existence for any finite family admitting a joint co-tile whose indicator is a fully periodic real function plus finitely many bounded real functions, each invariant under a subgroup of rank $d-1$. Extending the Boolean normal-form method of Greenfeld and Tao, we obtain a configuration in the translation orbit closure whose restrictions to cosets of a finite-index lattice are either invariant under a rank-$(d-1)$ subgroup or given by a three-term affine fractional-part formula with density one half. The tiling equations pair the restrictions given by this formula and determine a finite bipartite graph. A two-coloring replaces those restrictions by constants while preserving every tiling equation. We deduce a characterization of periodic tilability by independent companions and a decision algorithm for families containing an independent $(d-1)$-subfamily, with the dimension as part of the input.

math.CO↗

arXiv2026

Sharp Diameter Bounds for Nonnegative Cyclotomic Multiples

Let $N\ge2$ and let $p$ be its least prime divisor. We prove that every nonzero polynomial with nonnegative real coefficients divisible by $Φ_N$ has support diameter at least $(p-1)N/p$. Equality holds precisely for positive scalar multiples of monomial shifts of the $p$-term geometric sum $\sum_{j=0}^{p-1} X^{jN/p}$, thereby proving a conjecture of Steinberger. The proof turns cyclotomic divisibility into the vanishing of the first $p-1$ Fourier moments of a positive measure on the circle and then applies a classical extremal trigonometric polynomial. As a consequence, we establish the Coven--Meyerowitz diameter bound under their tiling conditions and determine its equality cases. Longer initial intervals of vanishing Fourier coefficients yield stronger diameter bounds, including an explicit refinement in terms of the prime-power divisor sets. The extremal trigonometric polynomial also yields a quantitative concentration estimate for measures and cyclotomic multiples with near-minimal support diameter.

math.NT↗

arXiv2026

Unsupported Cyclotomic Divisors in Three-Prime Integer Tilings

Cyclotomic divisibility imposes strong prime-power structure on integer tiles. We study unsupported cyclotomic divisors: mixed-order divisors for which none of the prime-power components of the order divides the mask, although every prime in the order divides the tile cardinality. Kiss, Łaba, Marshall and Somlai asked whether such a phenomenon can occur in the three-prime setting. We prove that unsupported cyclotomic divisors already occur for periods with three distinct prime factors. For primes $p<q<r$, we characterize the square-period case: an unsupported factor $Φ_{pqr}$ occurs in a tiling of $\ZZ_{(pqr)^2}$ if and only if $r\in\langle p,q\rangle$, and every such tile lies in a single residue class modulo $r$. Among cyclic tilings with the unsupported order dividing the specified modulus, the smallest modulus is $180$; if the order has three distinct prime factors, it is $900$. An Apéry-set construction gives examples for every triple at period $p^2q^2r^3$.The proof of our results combines Fourier rigidity, a three-cylinder decomposition, and an integer mass obstruction.

math.NT↗

arXiv2026

FADTI: Fourier and Attention Driven Diffusion for Multivariate Time Series Imputation

Multivariate time series imputation is fundamental in applications such as healthcare, traffic forecasting, and biological modeling, where sensor failures and irregular sampling lead to pervasive missing values. Existing Transformer- and diffusion-based imputers achieve strong performance, but they often rely mainly on time-domain modeling and lack adaptive spectral bias for recovering structured temporal gaps. We propose FADTI, a Fourier- and attention-driven diffusion framework for multivariate time series imputation. FADTI introduces a Fourier Bias Projection (FBP) module that injects learnable frequency-aware bias into intermediate hidden states during denoising. It projects intermediate hidden states onto Fourier bases, avoiding direct spectral estimation from masked or zero-filled inputs. With DFT, STFT, and FSST instantiations, FBP captures global periodicity, localized time--frequency variations, and non-stationary oscillatory patterns. By coupling FBP with self-attention and gated convolution, FADTI integrates frequency-domain guidance, temporal dependency modeling, and probabilistic denoising in a unified framework. Experiments on multiple benchmarks, including a new biological imputation benchmark, show that FADTI improves accuracy, uncertainty estimation, and sampling efficiency, especially under high missing rates and structured missing patterns. Code is available at https://github.com/RazeenLI/FADTI

cs.LG↗

arXiv2026

ISO-RAG: Isoperimetric Noise Control for Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) mitigates large language models (LLMs) hallucinations, yet conventional dense retrieval struggles with the complex reasoning paths of multi-hop question answering (QA). Graph-based RAG captures multi-step relationships but suffers from severe semantic drift and high online latency due to noisy global graph traversals. Thus, we propose ISO-RAG (ISOperimetric Retrieval-Augmented Generation), a geometry-aware RAG framework. By projecting the underlying knowledge graph into a hyperbolic Poincare ball to precompute node-wise isoperimetric profiles, ISO-RAG prunes spurious edges during retrieval, restricting the search space to a strictly localized subgraph. This topological purification regulates Personalized PageRank (PPR) diffusion driving the retrieval process, ensuring exact and low-latency convergence. Experiments on multi-hop QA benchmarks demonstrate that ISO-RAG outperforms state-of-the-art baselines by average absolute gains of 10.0% in retrieval recall and 4.3% in downstream exact match, achieving a superior accuracy-efficiency trade-off by fundamentally eliminating the latency bottleneck of global traversals. Our source code is available at https://github.com/ZaiizaiZHANG/ISO-RAG.

cs.AI↗

arXiv2026

Quantum information in neutron-proton scattering from the $M$ matrix

We study quantum-information aspects of neutron--proton scattering in the spin-space $M$-matrix framework. Four representative classes of input states are considered, namely diagonal mixed states, separable pure states, general two-qubit pure states, and a special Schmidt-like entangled subclass. For each class, ensemble-averaged output mutual information, reduced-state linear entropy, negativity, and geometric quantum discord are calculated in the relative momentum-- scattering angle plane. The results show that the outgoing spin correlations are governed jointly by scattering kinematics and by the structure of the incoming quantum ensemble. Input states with stronger intrinsic coherence or entanglement give larger maxima and higher minima in the mutual information, negativity, and geometric quantum discord. The enhanced regions of the mutual information and geometric discord depend on the input states, while the negativity maximum remains concentrated in the high-momentum backward-scattering region. These results extend earlier studies based on product-state entanglement power and provide an ensemble-based description of how spin correlations in neutron--proton scattering arise from the interplay between input-state structure and scattering dynamics.

nucl-th↗