Search arXiv⌕ Search

arXiv subjects

Yingying Zhang

Publications and source records attributed to Yingying Zhang.

At least 19 recordsLinked to original sources
arXiv2026

BEACON: Bridging Multimodal Semantics and Asynchronous Propagation for Cloud System Anomaly Detection

Modern large-scale cloud systems continuously generate massive and multimodal observability data, such as time-series metrics, log records, and system dependency graphs, which reflect the health state of the system. To ensure reliable cloud system operation, anomaly detection models identify abnormal states by analyzing these multimodal observations. However, existing approaches rely on simple fusion strategies and graph learning, ignoring the semantic gap among modalities and the asynchronous spatio-temporal dependencies across system instances over time. To this end, we propose BEACON, a framework to Bridge multimodal sEmantics and Asynchronous propagation for Cloud system anOmaly detectioN. Specifically, we introduce a Dynamic Interpretable Multimodal Aggregator (DIMA) that leverages an information-bottleneck mechanism to compress information from the semantics and temporal dynamics of metrics, and aligns them with logs in a shared prototype space. Furthermore, we propose an Asynchronous Spatio-Temporal Propagator (ASTP) that captures temporal dependencies with multi-scale temporal attention and models spatial relations with topology-aware aggregation. This enables asynchronous propagation of anomalies across system instances. Extensive experiments on three real-world cloud system datasets demonstrate that \textbf{BEACON} consistently outperforms state-of-the-art baselines in anomaly detection.

cs.LG↗
arXiv2026

Exploring Recurrent-Depth Language Models for Multitask Clinical Text Processing

Clinical natural language processing includes diverse tasks such as information extraction, classification, and question answering. Developing one model for all tasks could simplify deployment and improve knowledge transfer. We evaluated whether small recurrent-depth language models can meet this need through joint fine-tuning and compared their performance with standard transformers. We fine-tuned Ouro1.4B-Thinking, Ouro-2.6B-Thinking, HRM-Text-1B, Qwen3-1.7B, and Qwen3-8B on 9 datasets covering information extraction, classification, inference, and question answering, and evaluated them on four additional datasets excluded from training. Across these 13 datasets, Ouro-1.4B-Thinking achieved higher performances than Qwen3-1.7B on 11 datasets; Ouro-2.6BThinking was competitive with Qwen3-8B on most tasks. Ouro models showed the largest gains on inference and question answering, whereas HRM-Text-1B showed mixed results. These findings support further study of recurrent-depth models for clinical tasks through controlled experiments to isolate architectural effects and investigate how recurrent computation influences reasoning, knowledge acquisition, and generalization.

cs.CL↗
arXiv2026

Structural Properties of DomiRank Centrality: Symmetry Axioms, an Entropy Bound, and Spectral Interpolation

DomiRank is a node-importance algorithm for unweighted networks, defined by a dynamical model governed by a competition-strength parameter, a dominance threshold, and a decay rate. We establish three structural results. First, on symmetry. Automorphism-related vertices share the same DomiRank value, and a graph whose values are pairwise distinct must be asymmetric. The orbit count bounds the number of distinct values, and regular and vertex-transitive graphs admit closed forms. Second, on entropy. The maximum of the DomiRank entropy over connected graphs, log_2 N, is attained if and only if the graph is regular. The initial entropy derivative is the degree-log-degree edge covariance, making the decrease rigorous near vanishing competition for structurally disassortative graphs, and an assortative counterexample shows it does not extend to all graphs; a sufficient-condition theorem and a refined conjecture cover the rest. Third, on spectrum. The normalized distribution is sigma-invariant if and only if the degree vector is an adjacency-matrix eigenvector. The parameter sigma interpolates continuously between degree and smallest-eigenvector centrality within the resolvent (Bonacich-type) family, its sensitivity being a second-order suppression operator on the degree vector. Graph transformations are also characterized, including the m-product. Experiments on four real networks and synthetic families up to 10^4 vertices confirm the interpolation and exhibit monotone entropy decrease, continuous ranking interpolation, decaying positive fraction and ranking reversals; where the positive fraction has decayed far from one, the limit describes the solution vector rather than the induced ranking. DomiRank is thus a tunable complement to principal-eigenvector centrality. A companion study quantifies how much of its task-level edge over a degree baseline survives degree correction.

cs.SI↗
arXiv2026

Femtosecond all-optical polarization manipulation of harmonic generation in an anisotropic strongly coupled plasmon-epsilon-near-zero metasurface

Ultrafast manipulation of light polarization plays a key role in modern high-speed photonics and optoelectronic operations. However, most current polarization-active modulation schemes are restricted by a limited operating wavelength range (visible to infrared), narrow-band resonant fundamental wavelength working condition, and relatively long modulation time (> 1 picosecond). Here, by taking advantage of the transient anisotropic nonlinearity of optical strong coupling between the epsilon-near-zero mode of indium tin oxide (ITO) film and polarization-selective plasmonic resonance, we achieve femtosecond polarization manipulation of broadband light that spans ultraviolet (UV) to visible region by controlling harmonic generation of the signal light under resonant pumping. We demonstrate polarization elliptic rotation of 23.3° at 407 nm for third-harmonic generation (THG) and 55.0° at 610 nm for second-harmonic generation (SHG) simultaneously within ~600 fs. In addition, the strongly coupled system can also serve as a universal femtosecond polarization modulation source, imprinting a 32.9° polarization rotation in the UV region by converting the modulation information of the fundamental signal light to that of THG light via a uniform ITO film in a cascaded configuration. Our study establishes a new paradigm for UV-based, ultrahigh-speed broadband polarization photonics with direct implications for high-speed optical information processing and programmable nonlinear photonic devices.

physics.optics↗
arXiv2026

Link prediction in complex networks via fusing node centrality and local similarity indices: a fair-protocol reassessment and parameter design principles

We reassess a published fusion method under a fair protocol. Local similarity indices are cheap but score every node pair without common neighbors zero, and weighting them by a node centrality product, as proposed for PageRank, is not robust once the protocol is made comparable across indices; on our eight-network panel the reported improvement does not survive that protocol for six of the seven local indices. We study these issues in a unified piecewise framework that modulates the local score multiplicatively by a centrality product when local information suffices and uses a small-dose centrality product as a completion term when it does not. A scale analysis and a reassessment on eight real-world networks (five-fold cross-validation, exhaustive negative samples) trace this to the O(n^-2) scale of the PageRank product, whose modulation factor is then a near-identity map, and to its rank correlation with degree exceeding 0.89; control arms show scale calibration is necessary but not sufficient. The min-max normalized DomiRank product instead has an O(1) range and substantial non-degree content, and a grid search yields unified parameters (omega=5, c=0.1) under which DR-JC/HPI/SO significantly outperform their local baselines already at the dataset level (p=0.027 and 0.035), the remaining four at the fold level only; DR-RA attains an average AUC of 0.9257, above nine of ten comparison methods, whereas degree-only preferential attachment reaches only 0.7736. Mechanism decomposition shows the two mechanisms contribute differently across index families and should be dosed separately, yielding a safety boundary 0 < c <= 0.3, a plateau omega in [0.5, 12] and a near-critical plateau sigma in [0.7, 0.95]. Two negative results bound the claim: under a degree-corrected benchmark the gains over local baselines vanish and reverse in sign, and the AUC gains do not translate into top-L hit rate.

cs.SI↗
arXiv2026

OpenMobile-2: Building Versatile Mobile Agents with Scalable Environments and App-Native Tools

We introduce OpenMobile-2, a near-frontier mobile agent with fully open training environments and recipes. We make three key advances: (1) Diverse environments with simulated commercial apps: We build MobileGym++, featuring 35 realistic, functionally rich commercial-style apps with cross-app workflows, while preserving full controllability for reset and verification. Together with newly configured apps in Android emulators, this yields a diverse playground spanning over 110 apps. (2) Open training data at scale: Building on this foundation, we curate nearly 12K mobile interaction trajectories for supervised fine-tuning and the largest open collection of verifiable RL training data for mobile agents, comprising over 2K executable tasks with automatic rewards. (3) Hybrid GUI and app-native tool use: We explore an experimental mobile-use setting where apps expose selected functionalities as app-native tools alongside their GUIs, allowing agents to interleave GUI actions and tool calls within a task. We implement this setting in MobileGym++ with over 300 carefully scoped tools across 50+ apps. Additionally, we introduce MobileGym++ Bench for complex, long-horizon commercial mobile scenarios, supporting both GUI-only and hybrid GUI-tool evaluation on a shared task suite. OpenMobile-2 performs competitively across established benchmarks, including AndroidWorld (79.9) and MobileWorld (50.4), while showing promising transfer to real-device mobile use, nearly doubling SPA-Bench performance from 31.9 to 59.6. All resources are available at https://os-copilot.github.io/OpenMobile2-Home/.

cs.SE↗
arXiv2026

MoonGS: High-quality Representation of the Lunar Surface via Gaussian Splatting Using Robust Depth Features from Image Pairs

High-quality 3D reconstruction of lunar terrain from sparse rover images is indispensable for autonomous lunar exploration, but remains challenging because viewpoint overlap is insufficient, surface textures are weak, and data volume is limited. We propose MoonGS, the first feed-forward 3D Gaussian Splatting framework tailored to lunar scenes. Given only two input images, MoonGS predicts pixel-aligned Gaussian primitives in a single forward pass and renders photorealistic novel views without any per-scene optimization. MoonGS (i) adopts an adaptable backbone design that seamlessly integrates advanced vision foundation models to extract robust depth features; (ii) integrates semantic priors in two manners: merging semantic cues with visual features to refine Gaussian parameter estimation, and adopting a semantic ranking loss that regularizes background depth; and (iii) employs an entropy-guided heuristic resampling strategy to augment sparse observations by selecting the most informative distant viewpoints with negligible overhead. Experiments on the LuSNAR benchmark and our synthetic weak-texture MoonBlender dataset show that MoonGS surpasses state-of-the-art feed-forward NeRF/3DGS baselines by +4.9 dB PSNR, +0.29 SSIM, and 40\% lower LPIPS while maintaining sub-second inference. Furthermore, we validate the broad applicability of our framework by demonstrating that it effectively leverages state-of-the-art backbones, including VGGT, to significantly boost performance. Qualitative evaluations on Chang'e mission imagery also show the best visual quality among compared methods, indicating robustness on real lunar data. The source code and dataset are publicly available at https://github.com/InRobots/MoonBlender.

cs.CV↗
arXiv2026

ReLMem: Learning Recurrent Memory for Longitudinal EHR Modeling

Longitudinal electronic health record (EHR) modeling requires integrating new visits with an expanding patient history. Yet the continual accumulation of clinical information imposes increasing computational and memory costs on large language models (LLMs) when they process and retain complete patient histories. A practical alternative is visit-wise recurrent compression, which incorporates each incoming visit into a compact, continually updated patient memory. However, under a fixed memory budget, successive updates must integrate new information without progressively losing critical historical evidence needed to subsequent tasks. To address this challenge, we introduce Recurrent Longitudinal Memory (ReLMem), a framework that learns to maintain fixed-capacity patient memory for efficient downstream prediction with a frozen LLM. ReLMem equips this LLM with lightweight compression adapters to recurrently update the memory from its previous state and each incoming visit, without rereading earlier records. Specifically, we develop a multi-granularity optimization strategy to preserve task-relevant information throughout recurrent updates and support downstream prediction from the final memory. The intermediate supervision aligns attention outputs from compressed memory and the full history under identical queries, while prediction supervision minimizes cross-entropy with ground truth answers conditioned on the final memory. On EHR-based medication prediction, ReLMem approaches the F1 scores of full-history baseline while reducing average retained historical storage by 97.1%. Under the same memory budget, it improves macro- and micro-F1 over the strongest compressed-memory baseline by 4.66 and 4.75 percentage points, respectively. These results highlight the value of learning recurrent patient memory for efficient longitudinal EHR modeling.

cs.LG↗
arXiv2026

Improving Medical Calculation of LLMs with Embedded Coding

Large Language Models (LLMs) perform well on medical examinations and question-answering benchmarks, but remain unreliable on medical calculation tasks that require exact numerical outputs. These calculations support high-stakes decisions such as medication dosing, organ-function assessment, and prognostic scoring, for which even small errors can have serious clinical consequences. We introduce MedCode, a framework that improves medical calculation by training LLMs to generate embedded executable code. Given a clinical context, the model identifies the relevant calculator, extracts its input variables, and produces a script that delegates arithmetic operations to a deterministic interpreter. Executing the script returns the calculated value together with an explanation and the appropriate unit. We construct supervised fine-tuning (SFT) and preference datasets from the MedCalc benchmark and additionally curate a dataset for calculation tasks in Intensive Care Unit (ICU) scenarios. We further propose weighted Direct Preference Optimization (wDPO), which adaptively emphasizes preference pairs that are difficult for the model to distinguish. Experiments with LLaMA3-8B, Qwen2.5-7B, and Mistral-7B show absolute accuracy gains of 20--30 percentage points, demonstrating the effectiveness of embedded code generation for medical calculation.

cs.AI↗
arXiv2026

RULER: Instance-aware Rubric Rewards for SVG Generation

Generating Scalable Vector Graphics (SVG) code from natural-language instructions is an open-ended task without absolute visual ground truth, leaving both evaluation and policy optimization without a faithful signal. Scalar metrics (CLIP, Aesthetic) calibrated on natural images transfer poorly to stylized vector content, and reusing them as RL rewards triggers reward hacking. We address both limitations with rubric-based scoring. We first establish empirically that prompting a vision-language judge with a multi-axis rubric correlates with human judgments far better than scalar metrics, both across samples and within instructions. Building on this finding, we introduce RULER (Instance-aware Rubric Rewards for Reinforcement Learning), which converts each instruction into an instance-aware rubric of six items spanning semantic, visual, and stylistic axes; a judge VLM scores rendered rollouts item-by-item, and the weighted satisfactions form a fine-grained reward optimized via Group Relative Policy Optimization. Because the rubric is derived from text alone, RULER requires neither paired SVG ground truth nor human preference labels. On MMSVG-Illustration and MMSVG-Icon, RULER lifts the rubric score from 0.432/0.395 to 0.693/0.683, surpassing dedicated SVG specialists and matching the substantially larger DeepSeek-V3, with ablations identifying rubric design as the active lever for RL on open-ended SVG generation. The project page is available at https://hangyuran.github.io/RULER/.

cs.CV↗
arXiv2026

Support $τ$-tilting modules over Morita context algebras: A bilateral approximation approach

Let $k$ be a field and let $Λ=\left(\begin{smallmatrix}A&N\\M&B\end{smallmatrix}\right)_{ϕ,ψ}$ be a finite-dimensional Morita context algebra. We introduce a bilateral approximation construction which glues support $τ$-tilting modules over $A$ and $B$ by alternately correcting the two corner components through minimal approximations and pushouts. When this process terminates, it yields a support $τ$-tilting $Λ$-module with the prescribed componentwise torsion class. The one-sided case recovers Zhang's triangular-matrix construction, while the two-sided compatibility conditions give direct corner induction and, for radical-valued connecting maps, are also necessary, extending the Gao--Huang criterion. Examples show that the bilateral correction process can terminate even when neither one-sided compatibility condition is satisfied, while in other examples the process never terminates.

math.RT↗
arXiv2026

ExplainRoute: A Pre-Deployment Audit Framework for Non-Answer-Giving Programming Tutors

Programming tutors should support learners' own explanations rather than immediately providing model answers. We present ExplainRoute, a pre-deployment audit framework for non-answer-giving programming tutors. Given a code line and a learner explanation, it estimates the explanation state and selects one of two bounded responses: a Feynman-style self-explanation prompt or a Socratic scaffold. The framework exposes its state, strategy, cited code fragment, and leakage risk through a machine-checkable contract. Unlike benchmarks that rank tutors by fluency alone, ExplainRoute audits information boundaries, response polarity, failure closure, and the value of learner-explanation visibility before classroom deployment. We evaluate it offline on the 1,770-pair SelfCode corpus using a code-group split, with 443 pairs reserved in 11 untouched holdout groups. The evaluation compares direct answers, fixed open self-explanation, fixed Socratic scaffolding, adaptive routing, and an adaptive no-state ablation. Contract validity reaches 100% for all pedagogical conditions. Adaptive routing matches the frozen reference rule on 60.5% of records, with state macro-F1 of 0.238 (Open: 0.229; Socratic: 0.246), showing no reliable adaptive advantage. An independent language-model judge scores adaptive responses 4.516/5, outperforming the no-state ablation (2.819/5) but slightly below fixed open self-explanation (4.598/5) and Socratic scaffolding (4.658/5). A blinded rubric evaluation on a stratified 40-row subset confirms that visible learner explanations improve information value while adaptive routing does not outperform fixed strategies. The contribution is a validated audit protocol and a boundary finding, rather than evidence of improved learning, retention, or causal instructional effectiveness.

cs.IR↗
arXiv2026

Controllable Affective Generation via Latent Vector Steering

Large Language Models (LLMs) often produce emotionally flattened responses after alignment, limiting their effectiveness in affect-sensitive applications. In this paper, we propose EmoVec, a lightweight framework for controllable affective generation via latent vector steering. EmoVec extracts emotion-specific directions from paired neutral and emotion-conditioned responses using contrastive activation addition, and further refines them through task-specific debiasing and principal subspace removal. During inference, these vectors are injected into the final residual stream with static or scenario-adaptive scaling, enabling continuous control over emotional intensity without updating model weights. Experiments across three LLMs and eight emotions show that EmoVec consistently improves emotional salience while largely preserving semantic content, fluency, and coherence. Ablation studies and human evaluation further confirm the effectiveness of vector purification and adaptive scaling, establishing EmoVec as a practical inference-time method for affective control in deployed LLMs.

cs.CL↗
arXiv2026

Left Schur subcategories in recollements

Recently left Schur subcategories in a length abelian category were introduced by Enomoto, which unify torsion-free classes and wide subcategories. In this paper, we give a construction of left Schur subcategories in the recollements of length abelian categories. Moreover, we show that the construction restricts to wide subcategories and torsion-free classes. As a consequence, we obtain an explicit construction of cofinally closed monobricks in recollements. Finally, we apply the results to a special case of the MacPherson-Vilonen construction of abelian recollements and triangular matrix algebras.

math.RT↗
arXiv2026

Longitudinal Bayesian Learning of Continuous Disease Position across the Alzheimer's Disease Continuum

Alzheimer's disease (AD) progresses as a continuous biological process, whereas most existing neuroimaging-based artificial intelligence methods remain limited to discrete diagnosis or clinical score prediction from cross-sectional imaging. In this work, we propose Disease Continuum Positioning (DCP), a longitudinal Bayesian Learning framework that continuously estimates disease severity from longitudinal diffusion tensor imaging (DTI). Specifically, DCP models disease severity as a low-dimensional probabilistic latent variable by jointly integrating longitudinal observations with weak clinical supervision, from which the proposed Disease Continuum Score (DCS) is derived to quantify an individual's position along the Alzheimer's disease continuum together with its associated uncertainty. Extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort demonstrate that DCP consistently outperforms representative disease progression methods. More importantly, comprehensive validation analyses show that DCS accurately characterizes disease severity, exhibits strong clinical relevance, preserves longitudinal disease evolution, and predicts future disease conversion. These results suggest that DCS provides a quantitative imaging-derived representation for continuous assessment of Alzheimer's disease progression beyond conventional diagnostic labels and clinical scores.

cs.LG↗
arXiv2026

CT-PrepAgent: Bounded Policy and Controlled Execution for Adaptive CT Data Preparation

Heterogeneous computed tomography (CT) acquisitions and diverse downstream task requirements limit the transferability of fixed data preparation workflows across data sources and tasks. Existing approaches typically rely on manually designed or dataset-specific rules, making it difficult to accommodate changes in acquisition conditions and analytical objectives without manual intervention. Large language model (LLM)-based agents have shown promise for automating medical workflows, yet their potential for adaptive CT data preparation remains largely unexplored. To bridge this gap, we propose CT-PrepAgent, which enables adaptive CT data preparation through a bounded policy and controlled deterministic execution. Deterministic inspection constructs structured data--task profiles, from which a policy decides an eligible DICOM series or predefined preprocessing profile, while the controlled execution flow guards, resolves, executes, and verifies the decision with bounded recovery when enabled and safe quarantine otherwise. Across three public CT segmentation tasks, CT-PrepAgent derived data-task adaptive preprocessing decisions and achieved the highest macro-average Dice. On two private raw-DICOM cohorts, CT-PrepAgent increased verified output yield from 61.7\% to 70.0\% and yielded similar registration metrics on common verified outputs. Controlled fault and replay tests validate bounded recovery, safe quarantine, and policy-free replay under tested fault and drift settings.

cs.AI↗
arXiv2026

MedKGent: A Large Language Model Agent Framework for Constructing Temporally Evolving Medical Knowledge Graph

The rapid expansion of medical literature challenges the scalable structuring of domain knowledge. Knowledge Graphs (KGs) offer a solution, yet current construction methods lack generalizability and ignore the temporal dynamics of evolving knowledge. To address this, we introduce MedKGent, a Large Language Model (LLM) agent framework for building temporally evolving medical KGs. Using over 10 million PubMed abstracts from 1975 to 2023, MedKGent incrementally constructs a KG daily via two specialized agents. The Extractor Agent identifies knowledge triples and assigns confidence scores, while the Constructor Agent integrates these triples into a temporal graph, reinforcing recurring knowledge and resolving conflicts. The resulting KG contains 156,275 entities and 2,971,384 triples, making it, to our knowledge, the largest LLM-derived medical KG to date. Automated and expert assessments showed triple-validity rates approaching 90%. In downstream evaluations, MedKGent-KG significantly improved retrieval-augmented generation for five LLMs across seven medical question-answering benchmarks. Together, these results position MedKGent as a scalable and temporally aware infrastructure for medical knowledge representation and literature-grounded AI research.

cs.CL↗
arXiv2026

Stable (2+1)-dimensional soliton and breather molecules in a cold Rydberg atomic gas

We investigate the formation of stable (2+1)-dimensional spatial-domain optical soliton molecules and breather molecules in a gas of Rydberg atoms, highlighting the role of the nonlocal nonlinearity, which is generated by the electromagnetically induced transparency in the Rydberg medium. The setting supports diverse species of large-size polygonal soliton molecules, including rectangular and oblique rhombuses, checkerboard cells, and hexagons. The analysis identifies two distinct formation regimes. In the case of moderately nonlocality, the long-range interactions alone stabilize the soliton molecules in the static form. In contrast, in the strongly nonlocal regime, initially imposed rotation is required to generate a centrifugal force that counteracts the strong attraction, resulting in stably rotating soliton molecules. The rotation period can be controlled by adjusting the system parameters. Furthermore, appropriate initial velocities can induce inherent breathing dynamics in the solitons, leading to the formation of breather molecules. Tuning the initial velocity, one can control the evolution of soliton molecules and breather molecules and even realize their mutual conversion. Our study offers a new scheme for engineering soliton molecules and breather molecules, and suggests new possibilities for the design of data processing and transmission in optical systems.

physics.optics↗