arXiv subjects

Xianbin Cao

Publications and source records attributed to Xianbin Cao.

At least 19 recordsLinked to original sources

arXiv2026

Teaching Vision-Language-Action Models What to See and Where to Look

Vision-Language-Action (VLA) models have emerged as a promising paradigm for end-to-end autonomous driving. However, existing VLAs' training relies heavily on text-centric visual question answering and chain-of-thought reasoning data, which emphasizes linguistic reasoning rather than action-grounded planning. As a result, the learned representations capture semantic knowledge but lack spatial dependencies crucial for reliable trajectory prediction. We propose DriveTeach-VLA, a framework that explicitly teaches VLAs what to see and where to look. Driving-aware Vision Distillation (DVD) injects driving-specific perceptual priors into the vision encoder, while 2D Trajectory-Guided Prompts (2D-TGP) provide spatial conditioning aligned with feasible driving trajectories. Together, they form a vision-guided learning pipeline: what to see (DVD pretraining) - where to look (TGP-guided SFT) - how to act (TGP-guided GRPO). DriveTeach-VLA achieves the state-of-the-art performance on NAVSIM and nuScenes. Our code is available at: https://github.com/ShivaTeam/DriveTeach-VLA.

cs.CV↗

arXiv2026

Adaptive Plug-and-Play Channel Estimation with Consistency Models for MIMO Systems

This paper proposes a consistency-model-based channel estimation algorithm for multiple-input multiple-output (MIMO) systems. The proposed algorithm employs a consistency model (CM) to learn the angle-domain channel distribution and uses the trained CM as a plug-and-play (PnP) generative prior for MIMO channel estimation. The proposed algorithm alternates between a pilot-observation-based data-consistency update and a CM-prior-based denoising update. In addition, the proposed algorithm adaptively selects the penalty parameter according to residual energy and residual whiteness, and adjusts the CM denoising level according to the observed signal-to-noise ratio (SNR), thereby avoiding the performance degradation caused by fixed inference schedules under varying observation conditions. Simulation results show that the proposed algorithm not only reduces the number of inference steps by 50%--90, but also achieves high estimation accuracy and favorable cross-dataset performance.

eess.SP↗

arXiv2026

Multimodal Large Language Model Enabled Robust Beamforming for HAP Downlink Communications

Small changes in high altitude platform (HAP) attitude can cause significant deviations in HAP downlink beam directions, thereby severely degrading HAP downlink communication performance. In this paper, we develop a multimodal large language model (LLM) enabled beamforming framework to achieve robust HAP downlink communications.Specifically, we design a vision-language LLM (VL-LLM) that learns from multivariate flight telemetry to forecast short-term HAP attitudes under platform shaking and support delay-aware proactive beam steering.We design an offline forecast-error calibration procedure to obtain upper bounds on forecast errors and improve the reliability of proactive analog beam steering.Based on the attitude forecasts, we proactively update the analog beamformer and propose a QoS-driven beamforming and admission method with a lightweight feasibility-enforcement step to satisfy instantaneous transmit-power and QoS requirements.Simulation results indicate that the designed VL-LLM can accurately capture changes in the HAP attitude and the proposed beamforming method achieves a 22.1% higher user service ratio and a 12.5% higher sum-rate than representative baselines.The measured mean and p99 total latencies are 36.24 ms and 40.13 ms, respectively, supporting practical delay-aware deployment.

cs.NI↗

arXiv2026

Generative AI Agent Empowered Power Allocation for HAP Propulsion and Communication Systems

High altitude platforms (HAPs) are emerging as a key enabler for 6G coverage, yet limited energy must be split between propulsion and communications. Most prior HAP studies ignore propulsion power or rely on surrogates that miss hull-propeller interference, leading to misestimated communication power budgets and degraded beamforming. More importantly, HAP power allocation is intrinsically a multi-system, multidisciplinary problem in which aerodynamics, propulsion-system efficiency, and communication-system performance (quality of service (QoS) and energy efficiency (EE)) are tightly coupled.To address these challenges, this paper designs an interactive generative artificial intelligence (AI)-empowered HAP power allocation agent.By interacting with the AI agent, we develop an accurate propulsion power consumption model that takes into account the aerodynamic interference between the HAP's hull and the propeller. Assisted by the AI agent, we further formulate a HAP beamforming problem to improve user QoS and enhance the EE of the HAP communication system.This paper also proposes a QoS-enhanced energy-efficient (Q3E) beamforming algorithm to solve the formulated problem.Simulation results demonstrate the accuracy of the propulsion-power model and the effectiveness of the Q3E algorithm.

cs.NI↗

arXiv2026

Noise-Robust Tiny Object Localization with Flows

Despite significant advances in generic object detection, a persistent performance gap remains for tiny objects compared to normal-scale objects. We demonstrate that tiny objects are highly sensitive to annotation noise, where optimizing strict localization objectives risks noise overfitting. To address this, we propose Tiny Object Localization with Flows (TOLF), a noise-robust localization framework leveraging normalizing flows for flexible error modeling and uncertainty-guided optimization. Our method captures complex, non-Gaussian prediction distributions through flow-based error modeling, enabling robust learning under noisy supervision. An uncertainty-aware gradient modulation mechanism further suppresses learning from high-uncertainty, noise-prone samples, mitigating overfitting while stabilizing training. Extensive experiments across three datasets validate our approach's effectiveness. Especially, TOLF boosts the DINO baseline by 1.2% AP on the AI-TOD dataset.

cs.CV↗

arXiv2025

Adaptive Subarray Segmentation: A New Paradigm of Spatial Non-Stationary Near-Field Channel Estimation for XL-MIMO Systems

To address the complexities of spatial non-stationary (SnS) effects and spherical wave propagation in near-field channel estimation (CE) for extremely large-scale multiple-input multiple-output (XL-MIMO) systems, this paper proposes an SnS-aware CE framework based on adaptive subarray partitioning. We first investigate spherical wave propagation and various SnS characteristics and construct an SnS near-field channel model for XL-MIMO systems. Due to the limitations of uniform array partitioning in capturing SnS, we analyze the adverse effects of the non-ideal array segmentation (over- and under-segmentation) on CE accuracy. To counter these issues, we develop a dynamic hybrid beamforming-assisted power-based subarray segmentation paradigm (DHBF-PSSP), which integrates power measurements with a dynamic hybrid beamforming structure to enable joint subarray partitioning and decoupling. A power-adaptive subarray segmentation (PASS) algorithm leverages the statistical properties of power profiles, while subarray decoupling is achieved via a subarray segmentation-based sampling method (SS-SM) under radio frequency (RF) chain constraints. For subarray CE, we propose a subarray segmentation-based assorted block sparse Bayesian learning algorithm under the multiple measurement vectors framework (SS-ABSBL-MMV). This algorithm exploits angular-domain block sparsity under a discrete Fourier transform (DFT) codebook and inter-subcarrier structured sparsity. Simulation results confirm that the proposed framework outperforms existing methods in CE performance.

eess.SP↗

arXiv2025

Uncertainty-Aware Gradient Stabilization for Small Object Detection

Despite advances in generic object detection, there remains a performance gap in detecting small objects compared to normal-scale objects. We reveal that conventional object localization methods suffer from gradient instability in small objects due to sharper loss curvature, leading to a convergence challenge. To address the issue, we propose Uncertainty-Aware Gradient Stabilization (UGS), a framework that reformulates object localization as a classification task to stabilize gradients. UGS quantizes continuous labels into interval non-uniform discrete representations. Under a classification-based objective, the localization branch generates bounded and confidence-driven gradients, mitigating instability. Furthermore, UGS integrates an uncertainty minimization (UM) loss that reduces prediction variance and an uncertainty-guided refinement (UR) module that identifies and refines high-uncertainty regions via perturbations. Evaluated on four benchmarks, UGS consistently improves anchor-based, anchor-free, and leading small object detectors. Especially, UGS enhances DINO-5scale by 2.6 AP on VisDrone, surpassing previous state-of-the-art results.

cs.CV↗

arXiv2025

Survey on Near-Space Information Networks: Channel Modeling, Transmission, and Networking Perspectives

Near-space information networks (NSINs) composed of high-altitude platforms (HAPs) and high- and low-altitude unmanned aerial vehicles (UAVs) are a new regime for providing quick, robust, and cost-efficient sensing and communication services. Precipitated by innovations and breakthroughs in manufacturing, materials, communications, electronics, and control techniques, NSINs have been envisioned as an essential component of the emerging sixth-generation of mobile communication systems. This article reveals some critical issues needing to be tackled in NSINs through conducting experiments and discusses the latest advances in NSINs in the research areas of channel modeling, networking, and transmission from a forward-looking, comparative, and technical evolutionary perspective. In this article, we highlight the characteristics of NSINs and present the promising use cases of NSINs. The impact of airborne platforms' unstable movements on the phase delays of onboard antenna arrays with diverse structures is mathematically analyzed. The recent advances in HAP channel modeling are elaborated on, along with the significant differences between HAP and UAV channel modeling. A comprehensive review of the networking techniques of NSINs in network deployment, handoff management, and network management aspects is provided. Besides, the promising techniques and communication protocols of the physical (PHY) layer, medium access control (MAC) layer, network layer, and transport layer of NSINs for achieving efficient transmission over NSINs are reviewed, and we have conducted experiments with practical NSINs to verify the performance of some techniques. Finally, we outline some open issues and promising directions for NSINs deserved for future study and discuss the corresponding challenges.

cs.NI↗

arXiv2025

Prompt as Knowledge Bank: Boost Vision-language model via Structural Representation for zero-shot medical detection

Zero-shot medical detection can further improve detection performance without relying on annotated medical images even upon the fine-tuned model, showing great clinical value. Recent studies leverage grounded vision-language models (GLIP) to achieve this by using detailed disease descriptions as prompts for the target disease name during the inference phase. However, these methods typically treat prompts as equivalent context to the target name, making it difficult to assign specific disease knowledge based on visual information, leading to a coarse alignment between images and target descriptions. In this paper, we propose StructuralGLIP, which introduces an auxiliary branch to encode prompts into a latent knowledge bank layer-by-layer, enabling more context-aware and fine-grained alignment. Specifically, in each layer, we select highly similar features from both the image representation and the knowledge bank, forming structural representations that capture nuanced relationships between image patches and target descriptions. These features are then fused across modalities to further enhance detection performance. Extensive experiments demonstrate that StructuralGLIP achieves a +4.1\% AP improvement over prior state-of-the-art methods across seven zero-shot medical detection benchmarks, and consistently improves fine-tuned models by +3.2\% AP on endoscopy image datasets.

cs.CV↗

arXiv2024

Joint Source-Channel Optimization for UAV Video Coding and Transmission

This paper is concerned with unmanned aerial vehicle (UAV) video coding and transmission in scenarios such as emergency rescue and environmental monitoring. Unlike existing methods of modeling UAV video source coding and channel transmission separately, we investigate the joint source-channel optimization issue for video coding and transmission. Particularly, we design eight-dimensional delay-power-rate-distortion models in terms of source coding and channel transmission and characterize the correlation between video coding and transmission, with which a joint source-channel optimization problem is formulated. Its objective is to minimize end-to-end distortion and UAV power consumption by optimizing fine-grained parameters related to UAV video coding and transmission. This problem is confirmed to be a challenging sequential-decision and non-convex optimization problem. We therefore decompose it into a family of repeated optimization problems by Lyapunov optimization and design an approximate convex optimization scheme with provable performance guarantees to tackle these problems. Based on the theoretical transformation, we propose a Lyapunov repeated iteration (LyaRI) algorithm. Both objective and subjective experiments are conducted to comprehensively evaluate the performance of LyaRI. The results indicate that, compared to its counterparts, LyaRI achieves better video quality and stability performance, with a 47.74% reduction in the variance of the obtained encoding bitrate.

eess.SP↗

arXiv2024

DNN Task Assignment in UAV Networks: A Generative AI Enhanced Multi-Agent Reinforcement Learning Approach

Unmanned Aerial Vehicles (UAVs) possess high mobility and flexible deployment capabilities, prompting the development of UAVs for various application scenarios within the Internet of Things (IoT). The unique capabilities of UAVs give rise to increasingly critical and complex tasks in uncertain and potentially harsh environments. The substantial amount of data generated from these applications necessitates processing and analysis through deep neural networks (DNNs). However, UAVs encounter challenges due to their limited computing resources when managing DNN models. This paper presents a joint approach that combines multiple-agent reinforcement learning (MARL) and generative diffusion models (GDM) for assigning DNN tasks to a UAV swarm, aimed at reducing latency from task capture to result output. To address these challenges, we first consider the task size of the target area to be inspected and the shortest flying path as optimization constraints, employing a greedy algorithm to resolve the subproblem with a focus on minimizing the UAV's flying path and the overall system cost. In the second stage, we introduce a novel DNN task assignment algorithm, termed GDM-MADDPG, which utilizes the reverse denoising process of GDM to replace the actor network in multi-agent deep deterministic policy gradient (MADDPG). This approach generates specific DNN task assignment actions based on agents' observations in a dynamic environment. Simulation results indicate that our algorithm performs favorably compared to benchmarks in terms of path planning, Age of Information (AoI), energy consumption, and task load balancing.

cs.AI↗

arXiv2024

Connection Performance Modeling and Analysis of a Radiosonde Network in a Typhoon

This paper is concerned with the theoretical modeling and analysis of uplink connection performance of a radiosonde network deployed in a typhoon. Similar to existing works, the stochastic geometry theory is leveraged to derive the expression of the uplink connection probability (CP) of a radiosonde. Nevertheless, existing works assume that network nodes are spherically or uniformly distributed. Different from the existing works, this paper investigates two particular motion patterns of radiosondes in a typhoon, which significantly challenges the theoretical analysis. According to their particular motion patterns, this paper first separately models the distributions of horizontal and vertical distances from a radiosonde to its receiver. Secondly, this paper derives the closed-form expressions of cumulative distribution function (CDF) and probability density function (PDF) of a radiosonde's three-dimensional (3D) propagation distance to its receiver. Thirdly, this paper derives the analytical expression of the uplink CP for any radiosonde in the network. Finally, extensive numerical simulations are conducted to validate the theoretical analysis, and the influence of various network design parameters are comprehensively discussed. Simulation results show that when the signal-to-interference-noise ratio (SINR) threshold is below -35 dB, and the density of radiosondes remains under 0.01/km^3, the uplink CP approaches 26%, 39%, and 50% in three patterns.

cs.NI↗

arXiv2024

P4Q: Learning to Prompt for Quantization in Visual-language Models

Large-scale pre-trained Vision-Language Models (VLMs) have gained prominence in various visual and multimodal tasks, yet the deployment of VLMs on downstream application platforms remains challenging due to their prohibitive requirements of training samples and computing resources. Fine-tuning and quantization of VLMs can substantially reduce the sample and computation costs, which are in urgent need. There are two prevailing paradigms in quantization, Quantization-Aware Training (QAT) can effectively quantize large-scale VLMs but incur a huge training cost, while low-bit Post-Training Quantization (PTQ) suffers from a notable performance drop. We propose a method that balances fine-tuning and quantization named ``Prompt for Quantization'' (P4Q), in which we design a lightweight architecture to leverage contrastive loss supervision to enhance the recognition performance of a PTQ model. Our method can effectively reduce the gap between image features and text features caused by low-bit quantization, based on learnable prompts to reorganize textual representations and a low-bit adapter to realign the distributions of image and text features. We also introduce a distillation loss based on cosine similarity predictions to distill the quantized model using a full-precision teacher. Extensive experimental results demonstrate that our P4Q method outperforms prior arts, even achieving comparable results to its full-precision counterparts. For instance, our 8-bit P4Q can theoretically compress the CLIP-ViT/B-32 by 4 $\times$ while achieving 66.94\% Top-1 accuracy, outperforming the learnable prompt fine-tuned full-precision model by 2.24\% with negligible additional parameters on the ImageNet dataset.

cs.CV↗

arXiv2024

Pilot-Aided Joint Time Synchronization and Channel Estimation for OTFS

This letter proposes a pilot-aided joint time synchronization and channel estimation (JTSCE) algorithm for orthogonal time frequency space (OTFS) systems. Unlike existing algorithms, JTSCE employs a maximum length sequence (MLS) rather than an isolated signal as the pilot. Distinctively, JTSCE explores MLS's autocorrelation properties to estimate timing offset and channel delay taps. After obtaining delay taps, closed-form expressions of Doppler and channel gain for each propagation path are derived. Simulation results indicate that, compared to its counterpart, JTSCE achieves better bit error rate performance, close to that with perfect time synchronization and channel state information.

eess.SP↗

arXiv2024

Energy-Efficient URLLC Service Provision via a Near-Space Information Network

The integration of a near-space information network (NSIN) with the reconfigurable intelligent surface (RIS) is envisioned to significantly enhance the communication performance of future wireless communication systems by proactively altering wireless channels. This paper investigates the problem of deploying a RIS-integrated NSIN to provide energy-efficient, ultra-reliable and low-latency communications (URLLC) services. We mathematically formulate this problem as a resource optimization problem, aiming to maximize the effective throughput and minimize the system power consumption, subject to URLLC and physical resource constraints. The formulated problem is challenging in terms of accurate channel estimation, RIS phase alignment, theoretical analysis, and effective solution. We propose a joint resource allocation algorithm to handle these challenges. In this algorithm, we develop an accurate channel estimation approach by exploring message passing and optimize phase shifts of RIS reflecting elements to further increase the channel gain. Besides, we derive an analysis-friend expression of decoding error probability and decompose the problem into two-layered optimization problems by analyzing the monotonicity, which makes the formulated problem analytically tractable. Extensive simulations have been conducted to verify the performance of the proposed algorithm. Simulation results show that the proposed algorithm can achieve outstanding channel estimation performance and is more energy-efficient than diverse benchmark algorithms.

cs.IT↗

arXiv2024

Heterogeneous Generative Knowledge Distillation with Masked Image Modeling

Small CNN-based models usually require transferring knowledge from a large model before they are deployed in computationally resource-limited edge devices. Masked image modeling (MIM) methods achieve great success in various visual tasks but remain largely unexplored in knowledge distillation for heterogeneous deep models. The reason is mainly due to the significant discrepancy between the Transformer-based large model and the CNN-based small network. In this paper, we develop the first Heterogeneous Generative Knowledge Distillation (H-GKD) based on MIM, which can efficiently transfer knowledge from large Transformer models to small CNN-based models in a generative self-supervised fashion. Our method builds a bridge between Transformer-based models and CNNs by training a UNet-style student with sparse convolution, which can effectively mimic the visual representation inferred by a teacher over masked modeling. Our method is a simple yet effective learning paradigm to learn the visual representation and distribution of data from heterogeneous teacher models, which can be pre-trained using advanced generative methods. Extensive experiments show that it adapts well to various models and sizes, consistently achieving state-of-the-art performance in image classification, object detection, and semantic segmentation tasks. For example, in the Imagenet 1K dataset, H-GKD improves the accuracy of Resnet50 (sparse) from 76.98% to 80.01%.

cs.CV↗

arXiv2023

Representation Disparity-aware Distillation for 3D Object Detection

In this paper, we focus on developing knowledge distillation (KD) for compact 3D detectors. We observe that off-the-shelf KD methods manifest their efficacy only when the teacher model and student counterpart share similar intermediate feature representations. This might explain why they are less effective in building extreme-compact 3D detectors where significant representation disparity arises due primarily to the intrinsic sparsity and irregularity in 3D point clouds. This paper presents a novel representation disparity-aware distillation (RDD) method to address the representation disparity issue and reduce performance gap between compact students and over-parameterized teachers. This is accomplished by building our RDD from an innovative perspective of information bottleneck (IB), which can effectively minimize the disparity of proposal region pairs from student and teacher in features and logits. Extensive experiments are performed to demonstrate the superiority of our RDD over existing KD methods. For example, our RDD increases mAP of CP-Voxel-S to 57.1% on nuScenes dataset, which even surpasses teacher performance while taking up only 42% FLOPs.

cs.CV↗

arXiv2023

QoE-Driven Video Transmission: Energy-Efficient Multi-UAV Network Optimization

This paper is concerned with the issue of improving video subscribers' quality of experience (QoE) by deploying a multi-unmanned aerial vehicle (UAV) network. Different from existing works, we characterize subscribers' QoE by video bitrates, latency, and frame freezing and propose to improve their QoE by energy-efficiently and dynamically optimizing the multi-UAV network in terms of serving UAV selection, UAV trajectory, and UAV transmit power. The dynamic multi-UAV network optimization problem is formulated as a challenging sequential-decision problem with the goal of maximizing subscribers' QoE while minimizing the total network power consumption, subject to some physical resource constraints. We propose a novel network optimization algorithm to solve this challenging problem, in which a Lyapunov technique is first explored to decompose the sequential-decision problem into several repeatedly optimized sub-problems to avoid the curse of dimensionality. To solve the sub-problems, iterative and approximate optimization mechanisms with provable performance guarantees are then developed. Finally, we design extensive simulations to verify the effectiveness of the proposed algorithm. Simulation results show that the proposed algorithm can effectively improve the QoE of subscribers and is 66.75\% more energy-efficient than benchmarks.

cs.NI↗