arXiv subjects

Zhuoming Li

Publications and source records attributed to Zhuoming Li.

8 recordsLinked to original sources

arXiv2026

Tri-Hybrid Beamforming Design for DMA-Aided Secure ISAC Systems

This paper proposes a tri-hybrid beamforming scheme for secure integrated sensing and communication (ISAC) with a dynamic metasurface antenna (DMA) architecture, where the base station (BS) is capable of communicating with legitimate users and sensing the target. There is also an eavesdropper in the system intending to eavesdrop on the confidential information. The tri-hybrid beamforming design problem is formulated with the objective of maximizing the sensing signal-to-noise ratio (SNR) under the constraints of secrecy spectral efficiency (SSE), transmit power, and physical structure limitations. We first solve the problem and obtain the optimized fully-digital beamforming solution through successive convex approximation (SCA) and semidefinite relaxation (SDR) approaches. A triple alternating optimization scheme is then developed to iteratively optimize the digital, analog, and DMA beamformers, progressively approximating the fully-digital solution. Numerical results demonstrate that the proposed secure tri-hybrid beamforming design for DMA-aided ISAC improves the sensing SNR by approximately 3 dB compared to a tri-hybrid beamforming scheme with a fixed DMA electromagnetic design.

eess.SP↗

arXiv2026

Prime Focus Spectrograph on the Subaru Telescope: Overview of Science Operations

The paper presents the science operation framework for Prime Focus Spectrograph (PFS or 'Onohi'ula in its Hawaiian name) installed at the 8.2m Subaru Telescope on the summit of Maunakea. PFS is a massively multiplexed, wide-field, fiber-fed spectrograph covering 1.25 square degrees with 2386 science fibers. The instrument has been offered to the Subaru scientific community since March 2025. In order to fully exploit the unique capabilities of PFS, the Subaru Telescope has introduced a new, dedicated science operation framework for PFS. The default observing mode is queue observing, and multiple observing programs (in the same field) can be executed in the same exposure to achieve high observing efficiency. The quality of an exposure is based on the delivered signal-to-noise ratio and is quantified in terms of 'effective exposure time', and exposures are taken until the allocated 'fiber hours' for each program or target are achieved. The fiber hour is a new unit for observing time at Subaru; if we expose a fiber for 1 hour under the fiducial conditions, it is 1 fiber hour. Each observing program is granted the total fiber hours by the Time Allocation Committee. In addition to normal observing programs, which are selected through the standard peer-review process, there are two filler categories; community filler and observatory filler. As the names imply, the former is proposed by the community and the latter is prepared by the observatory. These filler targets are used whenever unassigned fibers are available. After an observing run, the data are fully reduced by the observatory and delivered to the users through the PFS Science Platform, a cloud-based data analysis environment. The paper gives a summary of all of this new framework and the actual implementation of it.

astro-ph.IM↗

arXiv2026

Segment-Aligned Policy Optimization for Multi-Modal Reasoning

Existing reinforcement learning approaches for Large Language Models typically perform policy optimization at the granularity of individual tokens or entire response sequences. However, such formulations often misalign with the natural step-wise structure of reasoning processes, leading to suboptimal credit assignment and unstable training in multi-modal reasoning tasks. To bridge this gap, we propose Segment-Aligned Policy Optimization (SAPO), a novel reinforcement learning paradigm that treats coherent reasoning steps, rather than tokens or full sequences as fundamental units of policy update. SAPO introduces a step-wise Markov decision process abstraction over reasoning segments, accompanied by segment-level value estimation, advantage computation, and importance sampling mechanisms that are semantically aligned with reasoning boundaries. Experiments on representative reasoning benchmarks demonstrate that SAPO consistently outperforms token-level and sequence-level policy optimization methods, achieving significant accuracy improvements while exhibiting better training stability and value estimation consistency. Our work underscores the importance of aligning reinforcement learning updates with the intrinsic structure of reasoning, paving the way for more efficient and semantically grounded policy optimization in complex reasoning tasks. Codes and models will be released to ensure full reproducibility.

cs.AI↗

arXiv2026

Dual Orthogonal Projections for Multiuser Interference Cancellation in mmWave Beamforming With Uniform Planar Arrays

This paper investigates multiuser interference (MUI) cancellation for millimeter-wave (mmWave) beamforming in communication systems. We propose a linear algorithm, termed iterative dual orthogonal projections (DOP), which alternates between two orthogonal projections: one to eliminate MUI and the other to refine combiners, ensuring empirical convergence in spectral efficiency. Simulation results show that, with each iteration, the spectral efficiency of each user converges rapidly, closely approaching the theoretical optimum determined by dirty paper coding (DPC), surpassing existing linear benchmarks while maintaining low computational complexity. Furthermore, the proposed DOP algorithm is extended to support both fully-digital and hybrid beamforming architectures.

eess.SP↗

arXiv2026

Attribute Distribution Modeling and Semantic-Visual Alignment for Generative Zero-shot Learning

Generative zero-shot learning (ZSL) synthesizes features for unseen classes, leveraging semantic conditions to transfer knowledge from seen classes. However, it also introduces two intrinsic challenges: (1) class-level attributes fails to capture instance-specific visual appearances due to substantial intra-class variability, thus causing the class-instance gap; (2) the substantial mismatch between semantic and visual feature distributions, manifested in inter-class correlations, gives rise to the semantic-visual domain gap. To address these challenges, we propose an Attribute Distribution Modeling and Semantic-Visual Alignment (ADiVA) approach, jointly modeling attribute distributions and performing explicit semantic-visual alignment. Specifically, our ADiVA consists of two modules: an Attribute Distribution Modeling (ADM) module that learns a transferable attribute distribution for each class and samples instance-level attributes for unseen classes, and a Visual-Guided Alignment (VGA) module that refines semantic representations to better reflect visual structures. Experiments on three widely used benchmark datasets demonstrate that ADiVA significantly outperforms state-of-the-art methods (e.g., achieving gains of 4.7% and 6.1% on AWA2 and SUN, respectively). Moreover, our approach can serve as a plugin to enhance existing generative ZSL methods.

cs.CV↗

arXiv2025

DiCaP: Distribution-Calibrated Pseudo-labeling for Semi-Supervised Multi-Label Learning

Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data in multi-label learning (MLL) by leveraging unlabeled data to improve the model's performance. While pseudo-labeling has become a dominant strategy in SSMLL, most existing methods assign equal weights to all pseudo-labels regardless of their quality, which can amplify the impact of noisy or uncertain predictions and degrade the overall performance. In this paper, we theoretically verify that the optimal weight for a pseudo-label should reflect its correctness likelihood. Empirically, we observe that on the same dataset, the correctness likelihood distribution of unlabeled data remains stable, even as the number of labeled training samples varies. Building on this insight, we propose Distribution-Calibrated Pseudo-labeling (DiCaP), a correctness-aware framework that estimates posterior precision to calibrate pseudo-label weights. We further introduce a dual-thresholding mechanism to separate confident and ambiguous regions: confident samples are pseudo-labeled and weighted accordingly, while ambiguous ones are explored by unsupervised contrastive learning. Experiments conducted on multiple benchmark datasets verify that our method achieves consistent improvements, surpassing state-of-the-art methods by up to 4.27%.

cs.LG↗

arXiv2025

Gestura: A LVLM-Powered System Bridging Motion and Semantics for Real-Time Free-Form Gesture Understanding

Free-form gesture understanding is highly appealing for human-computer interaction, as it liberates users from the constraints of predefined gesture categories. However, the sole existing solution GestureGPT suffers from limited recognition accuracy and slow response times. In this paper, we propose Gestura, an end-to-end system for free-form gesture understanding. Gestura harnesses a pre-trained Large Vision-Language Model (LVLM) to align the highly dynamic and diverse patterns of free-form gestures with high-level semantic concepts. To better capture subtle hand movements across different styles, we introduce a Landmark Processing Module that compensate for LVLMs' lack of fine-grained domain knowledge by embedding anatomical hand priors. Further, a Chain-of-Thought (CoT) reasoning strategy enables step-by-step semantic inference, transforming shallow knowledge into deep semantic understanding and significantly enhancing the model's ability to interpret ambiguous or unconventional gestures. Together, these components allow Gestura to achieve robust and adaptable free-form gesture comprehension. Additionally, we have developed the first open-source dataset for free-form gesture intention reasoning and understanding with over 300,000 annotated QA pairs.

cs.CV↗

arXiv2025

AREE-Based Decoupled Design of Hybrid Beamformers in mmWave XL-MIMO Systems

Hybrid beamforming has been widely employed in mmWave communications such as vehicular-to-everything (V2X) scenarios, as a compromise between hardware complexity and spectral efficiency. However, the inherent coupling between analog and digital precoders in hybrid array architecture significantly limits the computational and spectral efficiency of existing algorithms. To address this issue, we propose an alternating residual error elimination (AREE) algorithm, which decomposes the hybrid beamforming problem into two low-dimensional subproblems, each exhibiting a favorable matrix structure that enables effective decoupling of analog and digital precoders from the matrix product formulation. These subproblems iteratively eliminate each other's residual errors, driving the original problem toward the optimal hybrid beamforming performance. The proposed initialization ensures rapid convergence, while a low-complexity geometric channel SVD algorithm is developed by transforming the high-dimensional sparse channel into a low-dimensional equivalent, thereby simplifying the derivation of subproblems. Simulation results demonstrate that the AREE algorithm effectively decouples analog and digital precoders with low complexity, achieves fast convergence, and offers higher spectral efficiency than existing beamforming methods.

eess.SP↗