arXiv subjects

Xuancheng Zhu

Publications and source records attributed to Xuancheng Zhu.

7 recordsLinked to original sources

arXiv2026

A Material-Aware Channel Model for Efficient CKM Generation via Environment Reconstruction

Channel knowledge map (CKM) is a promising technology for environment-aware wireless communication, sensing, and localization in 6G networks. Accurate CKM generation requires precise reconstruction of the environment, including 3D geometries and scatterer materials, typically from multi-modal sensory observations such as LiDAR point clouds and sparse channel measurements. While the former is relatively easy to acquire, materials remain difficult to obtain directly from sparse channel measurements due to the lack of an explicit channel model linking them. To fill this gap, this paper proposes a material-aware channel model that explicitly characterizes the influence of scatterer materials on the wireless channel. Based on this model, an iterative gradient descent based material reconstruction algorithm is proposed. Full wave simulation results validate the developed model and the proposed algorithm, demonstrating their potentials for efficient CKM generation via environment reconstruction.

eess.SP↗

arXiv2026

SelPE: Progressive Selection for Private Structured Text Synthesis

Many data-driven applications rely on structured textual records, such as clinical triage notes and financial transaction logs, for downstream learning and decision-making. In privacy-sensitive domains, access to such records is strictly regulated, often resulting in only a small number of available private examples for model development and analysis. Yet existing differential privacy data synthesis methods fall short: tabular techniques cannot faithfully model free-form text, while text-based approaches often break structural constraints. We propose SelPE, a selection-guided progressive evolution framework for small-sample private structured text synthesis. Rather than relying on noisy aggregation or private model training, SelPE concentrates privacy budget on a sequence of multi-batch top-1 selections, enabling efficient guidance under tight privacy constraints. To support faithful and valid synthesis, SelPE decouples semantic abstraction from schema realization via a two-stage generation pipeline, and evaluates candidates using a multi-channel distance kernel that jointly models textual, categorical, and numeric fields in their native representations. A non-private contrastive expansion mechanism further promotes diversity without incurring additional privacy cost. Extensive Experiments demonstrate that SelPE consistently improves structural validity, fidelity, and downstream utility under strict differential privacy budgets, particularly in low-data regimes.

cs.CR↗

arXiv2026

AutoRAS: Learning Robust Agentic Systems with Primitive Representations

The automated design of agentic systems offers a promising pathway for scaling large language models (LLMs) beyond single-agent reasoning. While prior work has advanced task performance through handcrafted or automatically generated multi-agent workflows, robustness is often treated as an afterthought, leaving systems vulnerable to external adversaries and internal failures. We propose AutoRAS, a framework for the Automated design of Robust Agentic Systems. AutoRAS formulates system design as generating a sequence of symbolic primitives that jointly encode structural connectivity and behavioral actions, and learns to optimize this sequence using execution-derived safety signals and flow-based sequence-level objectives. Extensive experiments show that AutoRAS achieves the best performance in both vanilla and adversarial settings, with the smallest performance degradation under attacks. Further analyses demonstrate strong transferability, stable optimization behavior, stability across primitive sets, and favorable cost trade-offs. Our code is available at $\href{https://github.com/guohezuy/AutoRAS}{\text{this https URL}}$.

cs.AI↗

arXiv2026

See First, Answer Later: Visual Evidence Pre-Alignment via Sufficiency-Driven RL

Multimodal large language models (MLLMs) integrate strong text reasoning with visual inputs, yet their responses can be inconsistent with the underlying images, indicating ineffective utilization of visual evidence during inference. The prevailing training paradigm relies on large-scale caption-based pretraining for general alignment, followed by supervised fine-tuning and reinforcement learning to enable instruction following and complex reasoning. However, such pretraining provides only weak visual grounding: short, coarse captions bias models toward salient objects while neglecting fine-grained visual evidence. In this paper, we introduce Visual Evidence Pre-Alignment (VEPA), an intermediate stage between pretraining and post-training that explores a novel sufficiency-driven objective with Group Relative Policy Optimization (GRPO) to optimize question-conditioned visual evidence descriptions. Extensive experiments across diverse benchmarks show that our VEPA consistently enhances performance on visually demanding evaluations and complements standard supervised post-training. Further analyses show that the income stems from strengthened, transferable visual grounding, rather than from additional task-specific training.

cs.CV↗

arXiv2026

Can LLM Agents Sustain Long-Horizon Organizational Dynamics?

Large language agents are increasingly used for social simulation, yet it remains unclear whether they can sustain coherent behavior in structured organizations, where goals must propagate through hierarchy, tasks depend on prior execution, and artifacts accumulate over long horizons. We formulate long-horizon organizational simulation as a memory-centered coordination problem and introduce TaskWeave, a hierarchical agentic framework that maintains planning states through a Formulate-Partition-Diagnose-Align cycle and grounds execution through dependency-aware trace memory. We evaluate TaskWeave in a year-long IT company simulation and compare it with other multi-agent frameworks on organizational coherence, execution grounding, and downstream enterprise NLP utility. Experiments show that TaskWeave supports coherent and long-horizon organizational dynamics while producing grounded artifacts and adapting to external environments. These findings suggest that structured simulation memory is a key mechanism for building reliable LLM-based organizational simulators.

cs.AI↗

arXiv2025

Cost-Effective XL-MIMO Communication with Cylinder Directly-Connected Antenna Array

Extremely large-scale multi-input multi-output (XL-MIMO) is a promising technology for the sixth generation (6G) wireless networks, thanks to its superior spatial resolution and beamforming gains. In order to realize XL-MIMO costeffectively, an innovative ray antenna array (RAA) architecture with directly-connected uniform linear array (ULA) was recently proposed, which achieves flexible beamforming without relying on traditional analog phase shifters or digital beamforming. However, RAA suffers from the signal blockage issue since its ray-configured ULAs are placed in the same plane. To address this issue, this paper proposes a novel antenna array architecture termed cylinder directly-connected antenna array (DCAA), which is achieved via multiple simple uniform circular array (sUCA) with carefully designed orientations in a layered three-dimensional structure. The so-called sUCA partitions the uniform circular array (UCA) into two sub-arrays where each sub-array has all antenna elements directly connected to achieve a desired beam direction corresponding to the sub-array's physical orientation, thus achieving full spatial coverage. Compared with the conventional ULA architecture with hybrid analog/digital beamforming (HBF), the proposed cylinder DCAA can achieve uniform spatial resolution, enhanced communication rate and lower hardware costs. Simulation results are provided to validate the promised gains of cylinder DCAA, demonstrating its great potential for high-frequency systems such as millimeter wave (mmWave) and Terahertz (THz) systems.

eess.SP↗

arXiv2025

Full-Angle Ray Antenna Array and Omnicell Wireless Communication System

Ray antenna array (RAA) was recently proposed as a novel multi-antenna architecture that arranges multiple massive cheap antenna elements into simple uniform linear arrays (sULAs) with different orientations. Compared with traditional architectures like hybrid analog/digital beamforming with uniform linear array (ULA) and uniform circular array (UCA), RAA has several promising advantages such as significantly reduced hardware cost, higher beamforming gains and the ability of providing uniform angular resolution for all directions. In this paper, we propose a full-angle RAA architecture and an innovative omnicell wireless communication paradigm enabled by full-angle RAA. The proposed full-angle RAA expands RAA's orientation angle to the full angle domain, such that the RAA's advantages can be exploited to all directions. This further enables the new concept of omnicell wireless communication system, with the base station equipped by full-angle RAA and deployed at the center of each cell. Compared to the conventional cell sectoring wireless communication system, the proposed omnicell system is expected to not only significantly reduce the inter-user interference, but also improve the cost efficiency. Extensive analytical and numerical results are provided to compare those key performance indicators such as the spatial resolution and the communication rate of the proposed full-angle RAA based omnicell wireless communication system against the conventional ULA/UCA-based cell sectoring systems.

eess.SP↗