arXiv subjects

Jian Xiao

Publications and source records attributed to Jian Xiao.

At least 19 recordsLinked to original sources

arXiv2026

SCALE: Real-Time Latent Communication for Cyber-Physical Agents

Latent communication enables agents to collaborate efficiently by directly exchanging intermediate model representations rather than raw sensor data or text messages. Existing approaches, however, are primarily designed for LLM agents and do not consider the bandwidth constraints and heterogeneous task deadlines faced by cyber-physical agents. In this work, we propose SCALE, a real-time latent communication method tailored for cyber-physical agents. Motivated by the spatio-temporal continuity of the physical world, SCALE matches tokens across consecutive latent messages using similarity scores to identify reusable information and avoid redundant transmission. It then profiles the semantic gain and communication cost of transmitting different amounts of latent information. Based on these profiles, SCALE employs a real-time scheduler that jointly determines which pending messages to transmit and how many latent tokens to send, subject to available bandwidth and task deadlines. We evaluate SCALE across three cyber-physical benchmarks and a physical spider robot case study under diverse network conditions and heterogeneous task deadlines. Our results show that SCALE significantly improves both task performance and on-time delivery in bandwidth-constrained networks. Our code is available at: https://anonymous.4open.science/r/SCALE-4E80/.

cs.RO↗

arXiv2026

Flow and transport characteristics of hydrogen in a square-spiral Knudsen pump with coupled thermal transpiration and Poiseuille flow

Gas transport driven by thermal transpiration under temperature gradients, known as a Knudsen pump, has shown strong potential in gas separation and delivery applications. However, its performance and operational stability are strongly affected by structural design, system integration, and manufacturing complexity, highlighting the need for efficient and manufacturable multi-stage configurations. In this study, a novel spiral-structured Knudsen pump is proposed. Within the microchannels, a temperature gradient drives gas from the low- to high-temperature region, inducing pressure rise and thermal transpiration. In the main channel, a pressure gradient drives gas from the high- to low-temperature region, forming Poiseuille flow. Numerical methods are employed to investigate the flow mechanisms, and the effects of varying the numbers of series and parallel microchannels on gas transmission behavior are evaluated. The results indicate that bends of the main channel lead to nonlinear variations in pressure and temperature, as well as flow deceleration in the curved regions. Increasing the number of serial channels reduces the pressure load across the microchannel, thereby enhancing the overall transport performance. For instance, a ten-stage serial configuration improves performance by 3.35% compared with the four-stage case, reaching 3.10E10-7 kg/(m*s). In contrast, increasing the number of parallel microchannels reduces pressure loading while enhancing gas rarefaction, thereby strengthening thermal transpiration. A five-stage parallel configuration yields an 70.80% improvement over the two-stage case, reaching 3.07E10-7 kg/(m*s). These results provide valuable guidance for the design and fabrication of high-performance Knudsen pumps.

physics.flu-dyn↗

arXiv2026

Correlation inequalities for volume polynomials

We establish sharp correlation inequalities for volume polynomials associated with covers of multisets. These inequalities simultaneously generalize the Khovanskii--Teissier and reverse Khovanskii--Teissier inequalities for intersections of nef divisor classes on projective varieties. They also hold for volume polynomials arising from convex bodies and Hermitian matrices, recovering and extending classical results such as the Alexandrov--Fenchel inequality for mixed volumes of convex bodies and the Hadamard--Fischer inequality for determinants of positive semidefinite matrices. As a combinatorial application, we prove a negative correlation inequality for partitions of bases of matroids that are linearly realizable over a field.

math.AG↗

arXiv2026

Not Where You Think: Protecting Location Privacy Against Reasoning-Based Geolocation

Recent large reasoning models can infer where ordinary photos were taken using only visual information, creating a serious location privacy risk even when metadata and obvious geographic identifiers are removed. To understand the mechanism behind such fine-grained localization, we develop GeoTrace, a structured reasoning attack that exposes how models extract location-relevant visual clues and compose them through coarse-to-fine geographic reasoning. We evaluate GeoTrace on a worldwide benchmark of 525 ordinary images spanning natural, urban, residential, and human-centered scenes while excluding famous landmarks and direct location indicators. Our results show that GeoTrace localizes 25.71% of images within 1 km with Gemini-3-Flash and 16.95% with GPT-5-mini. More importantly, our analysis reveals a key insight in the visual clue chain: broad scene properties mainly support coarse localization, whereas precise localization disproportionately relies on small, distinctive, and editable local clues. Building on this insight, we develop GeoShift, a proactive user-side defense that breaks the visual clue chain before image release. Rather than simply removing geographic evidence, GeoShift preserves stable scene properties, selects a visually compatible false target location, and strategically rewrites high-impact geo-revealing clues to construct a coherent alternative geographic hypothesis. Experiments show that GeoShift reduces street-level localization from 16.95% to 2.48% under GPT-5-mini and from 25.71% to 3.81% under Gemini-3-Flash, while preserving visual quality and remaining effective against adaptive attacks.

cs.CR↗

arXiv2026

Cold-Atom Rydberg Array with Spatiotemporal Multiplexing for High-Sensitivity Wireless Communications

An effective Cold-Atom Rydberg Array (CoRA) architecture is investigated, which addresses the fundamental thermal noise limit in classical electromagnetics and intrinsic Doppler broadening as well as photon shot noise in conventional hot-vapor cell-based atomic receivers. However, translating the CoRA-based quantum electrometers into practical wireless communication receivers is bottlenecked by both the macroscopic readout dead-times and nonlinear binomial population counts rather than the additive-Gaussian baseband samples in conventional receivers. To address these critical challenges and facilitate the practicality of the CoRA, in this paper, a subarray scheduling-based spatiotemporal multiplexing (STM) framework is first proposed for gap-free symbol interrogation within an admitted packet. Additionally, a signal detection scheme based on a joint binomial particle smoother (JBPS) is proposed for burst-wise state and data inference. Furthermore, the weak-signal quantum-projection-noise (QPN) limit and the state-averaged finite-alphabet rate are derived for CoRA-assisted wireless communications. Numerical results demonstrate that: 1) The proposed STM-based CoRA scheme resolves the macroscopic temporal bottleneck for continuous signal sampling; 2) The proposed JBPS approach effectively mitigates the nonlinear noise amplification to achieve a lower bit error rate; and 3) Performance analysis reveals a fundamental sensitivity-sustainability tradeoff that increasing the number of atoms effectively suppresses the QPN, but simultaneously reduces the sustainable packet repetition rate.

cs.IT↗

arXiv2026

Rydberg Atomic Quantum Receivers for Wireless Communications: Two-Color vs. Three-Color Excitation

An efficient three-color (3C) laser excitation-based Rydberg atomic quantum receiver (RAQR) architecture is investigated for wireless communications, utilizing a five-level (5L) electronic transition mechanism. Specifically, the conventional two-color (2C) RAQR with the four-level (4L) excitation faces three fundamental obstacles: 1) high cost and engineering challenges due to the reliance on unstable short-wavelength lasers; 2) a fundamental sensitivity limit in thermal atoms caused by residual Doppler broadening; and 3) the inability to detect low-frequency bands due to the energy-level constraint of two-photon resonance. To address these challenges, this paper analyzes a 3C5L-RAQR architecture with all-red/infrared lasers, which not only solves the engineering cost issues but also enables effective Doppler cancellation and low-frequency detection by exploiting the three-photon resonance. Bridging atomic physics and communication theory, an end-to-end equivalent baseband signal model is derived. Furthermore, the performance of different RAQR architectures is evaluated in terms of sensitivity, achievable rate and spectrum access range. Moreover, we provide an exact numerical solution for practical RAQRs by employing the Liouvillian superoperator formalism. Numerical results demonstrate that the exhibited 3C5L-RAQR achieves superior sensitivity compared to the conventional 2C4L-RAQR and a classical antenna-based radio frequency receiver for weak-signal detection. Finally, the inherent sensitivity-rate trade-off is revealed, showing that the 3C5L-RAQR is more suitable for deployment in power-limited communication scenarios demanding broad spectrum access.

cs.IT↗

arXiv2026

GenCOPE: Syn2Real Generalized Category-Level Object Pose Estimation for Robotic Picking

Category-level object pose estimation (COPE), capable of generalizing to intra-class unknown objects, has become a core technique for robotic 3D scene understanding. However, existing COPE methods still require labor-intensive recollection of real-world training data for novel object categories, which limits their scalability in practical applications. This paper aims to achieve synthetic-to-real (Syn2Real) generalized COPE, where a model is trained solely on rendered synthetic data and directly generalized to real-world deployments. The central challenge lies in the significant domain gap between synthetic and real-world data, particularly in texture appearance. To address this, we aim to enhance domain generalization by learning domain-invariant representations that capture semantic commonalities among objects within the same category. We introduce 2D and 3D semantic consistency constraints to reduce the sensitivity of feature encoders to domain-specific features. In addition, we propose an end-to-end pose regression framework that performs 2D-3D cross consistency learning, leveraging dense cross-modality fusion to further refine pose estimation. Since simplicity and effectiveness are essential for real-world robotic deployment, our model operates exclusively on global features, yielding a highly lightweight and efficient architecture. Extensive experiments on the REAL275 and Wild6D benchmarks, as well as real-world robotic manipulation scenes, show superior Syn2Real generalization performance of our paradigm. Code and demos are released at https://paperreview99.github.io/GenCOPE/.

cs.CV↗

arXiv2026

Shift-MoE-Based DJSCC for CSI Feedback in Multi-User Pinching-Antenna Systems

In frequency-division duplexing systems, the performance gains of pinching-antenna systems (PASS) critically depend on accurate channel state information (CSI) at the base station. However, PASS CSI exhibits structured correlations over the waveguide-antenna grid and pronounced heterogeneity across users, making conventional fixed feedback mappings difficult to generalize. To address this challenge, this letter proposes an end-to-end CSI feedback scheme over a noisy uplink feedback link based on deep joint source-channel coding, termed Shift-based Mixture-of-Experts (Shift-MoE). Specifically, Shift-MoE leverages channel-grouped one-step shift operations to capture grid dependencies without global attention, and employs a gated multilayer perceptron mixture-of-experts module to adapt to heterogeneous CSI statistics across users. Numerical results demonstrate that the proposed Shift-MoE consistently outperforms representative learning-based CSI feedback baselines in normalized mean squared error and remains effective under different system parameter settings.

eess.SP↗

arXiv2026

Occlusion-Aware 3D Hand-Object Pose Estimation with Masked AutoEncoders

Hand-object pose estimation from monocular RGB images remains a significant challenge mainly due to the severe occlusions inherent in hand-object interactions. Existing methods do not sufficiently explore global structural perception and reasoning, which limits their effectiveness in handling occluded hand-object interactions. To address this challenge, we propose an occlusion-aware hand-object pose estimation method based on masked autoencoders, termed as HOMAE. Specifically, we propose a target-focused masking strategy that imposes structured occlusion on regions of hand-object interaction, encouraging the model to learn context-aware features and reason about the occluded structures. We further integrate multi-scale features extracted from the decoder to predict a signed distance field (SDF), capturing both global context and fine-grained geometry. To enhance geometric perception, we combine the implicit SDF with an explicit point cloud derived from the SDF, leveraging the complementary strengths of both representations. This fusion enables more robust handling of occluded regions by combining the global context from the SDF with the precise local geometry provided by the point cloud. Extensive experiments on challenging DexYCB and HO3Dv2 benchmarks demonstrate that HOMAE achieves state-of-the-art performance in hand-object pose estimation. We will release our code and model.

cs.CV↗

arXiv2026

GenHOI: Generalized Hand-Object Pose Estimation with Occlusion Awareness

Generalized 3D hand-object pose estimation from a single RGB image remains challenging due to the large variations in object appearances and interaction patterns, especially under heavy occlusion. We propose GenHOI, a framework for generalized hand-object pose estimation with occlusion awareness. GenHOI integrates hierarchical semantic knowledge with hand priors to enhance model generalization under challenging occlusion conditions. Specifically, we introduce a hierarchical semantic prompt that encodes object states, hand configurations, and interaction patterns via textual descriptions. This enables the model to learn abstract high-level representations of hand-object interactions for generalization to unseen objects and novel interactions while compensating for missing or ambiguous visual cues. To enable robust occlusion reasoning, we adopt a multi-modal masked modeling strategy over RGB images, predicted point clouds, and textual descriptions. Moreover, we leverage hand priors as stable spatial references to extract implicit interaction constraints. This allows reliable pose inference even under significant variations in object shapes and interaction patterns. Extensive experiments on the challenging DexYCB and HO3Dv2 benchmarks demonstrate that our method achieves state-of-the-art performance in hand-object pose estimation.

cs.CV↗

arXiv2026

Numerical characterization of the hard Lefschetz classes of dimension two, I: supercritical collections under certain rearrangement

We study the numerical characterization of two dimensional hard Lefschetz classes given by the complete intersections of nef classes. In Shenfeld and van Handel's breakthrough work on the characterization of the extremals of the Alexandrov-Fenchel inequality for convex polytopes, they proposed an open question on the algebraic analogue of the characterization. We settle the open question when the collection of nef classes is given by a rearrangement of supercriticality, which in particular includes the big nef collection as a special case. The main results enable us to refine some previous results and study the extremals of Hodge index inequality, and also provide the first series of examples of hard Lefschetz classes of dimension two both in algebraic geometry and analytic geometry, in which one can allow nontrivial augmented base locus and thus drop the semi-ampleness or semi-positivity assumption. As a key ingredient of the numerical characterization, we establish a local Hodge index inequality for Lorentzian polynomials, which is the algebraic analogue of the local Alexandrov-Fenchel inequality obtained by Shenfeld-van Handel for convex polytopes. This result holds in broad contexts, e.g., it holds on a smooth projective variety, on a compact Kähler manifold, and on a Lorentzian fan, which contains the Bergman fan of a matroid or a polymatroid as a typical example.

math.AG↗

arXiv2026

Channel Estimation for Rydberg Atomic Quantum Receivers: Unrolled Phase Retrieval from Holographic Snapshots

A model-driven deep learning framework is proposed for channel estimation in Rydberg atomic quantum receivers (RAQRs) based on the measurement of holographic snapshots. Specifically, we develop a Transformer-based unrolling architecture, termed URformer, to solve the non-linear biased phase retrieval problem, which is derived by unrolling a stabilized variant of the expectation-maximization Gerchberg-Saxton (EM-GS) algorithm. Each layer of the proposed URformer incorporates three trainable modules: 1) a learnable filter network that replaces the fixed Bessel kernel in the classic EM-GS algorithm; 2) a trainable gating mechanism that adaptively combines classic updates to ensure training stability; and 3) an efficient channel Transformer module that learns to correct residual errors by capturing non-local channel dependencies. Numerical results demonstrate that the proposed URformer significantly outperforms classic iterative algorithms and conventional black-box neural networks with less pilot overhead.

cs.IT↗

arXiv2026

Channel Estimation for Flexible Intelligent Metasurfaces: From Model-Based Approaches to Neural Operators

Flexible intelligent metasurfaces (FIMs) offer a new solution for wireless communications by introducing morphological degrees of freedom, dynamically morphing their three-dimensional shape to ensure multipath signals interfere constructively. However, realizing the desired performance gains in FIM systems is critically dependent on acquiring accurate channel state information across a continuous and high-dimensional deformation space. Therefore, this paper investigates this fundamental channel estimation problem for FIM assisted millimeter-wave communication systems. First, we develop model-based frameworks that structure the problem as either function approximation using interpolation and kernel methods or as a sparse signal recovery problem that leverages the inherent angular sparsity of millimeter-wave channels. To further advance the estimation capability beyond explicit assumptions in model-based channel estimation frameworks, we propose a deep learning-based framework using a Fourier neural operator (FNO). By parameterizing a global convolution operator in the Fourier domain, we design an efficient FNO architecture to learn the continuous operator that maps FIM shapes to channel responses with mesh-independent properties. Furthermore, we exploit a hierarchical FNO (H-FNO) architecture to efficiently capture the multi-scale features across a hierarchy of spatial resolutions. Numerical results demonstrate that the proposed H-FNO significantly outperforms the model-based benchmarks in estimation accuracy and pilot efficiency. In particular, the interpretability analysis show that the proposed H-FNO learns an anisotropic spatial filter adapted to the physical geometry of FIM and is capable of accurately reconstructing the non-linear channel response across the continuous deformation space.

cs.IT↗

arXiv2026

Wireless AI Evolution: From Statistical Learners to Electromagnetic-Guided Foundation Models

While initial applications of artificial intelligence (AI) in wireless communications over the past decade have demonstrated considerable potential using specialized models for targeted communication tasks, the revolutionary demands of sixth-generation (6G) networks for holographic communications, ubiquitous sensing, and native intelligence are propelling a necessary evolution towards AI-native wireless networks. The arrival of large AI models paves the way for the next phase of Wireless AI, driven by wireless foundation models (WFMs). In particular, pre-training on universal electromagnetic (EM) principles equips WFMs with the essential adaptability for a multitude of demanding 6G applications. However, existing large AI models face critical limitations, including pre-training strategies disconnected from EM-compliant constraints leading to physically inconsistent predictions, a lack of embedded understanding of wave propagation physics, and the inaccessibility of massive labeled datasets for comprehensive EM-aware training. To address these challenges, this article presents an electromagnetic information theory-guided self-supervised pre-training (EIT-SPT) framework designed to systematically inject EM physics into WFMs. The EIT-SPT framework aims to infuse WFMs with intrinsic EM knowledge, thereby enhancing their physical consistency, generalization capabilities across varied EM landscapes, and overall data efficiency. Building upon the proposed EIT-SPT framework, this article first elaborates on diverse potential applications in 6G scenarios of WFMs, then validates the efficacy of the proposed framework through illustrative case studies, and finally summarizes critical open research challenges and future directions for WFMs.

cs.IT↗

arXiv2026

Efficient Cross-Architecture Knowledge Transfer for Large-Scale Online User Response Prediction

Deploying new architectures in large-scale user response prediction systems incurs high model switching costs due to expensive retraining on massive historical data and performance degradation under data retention constraints. Existing knowledge distillation methods struggle with architectural heterogeneity and the prohibitive cost of transferring large embedding tables. We propose CrossAdapt, a two-stage framework for efficient cross-architecture knowledge transfer. The offline stage enables rapid embedding transfer via dimension-adaptive projections without iterative training, combined with progressive network distillation and strategic sampling to reduce computational cost. The online stage introduces asymmetric co-distillation, where students update frequently while teachers update infrequently, together with a distribution-aware adaptation mechanism that dynamically balances historical knowledge preservation and fast adaptation to evolving data. Experiments on three public datasets show that CrossAdapt achieves 0.27-0.43% AUC improvements while reducing training time by 43-71%. Large-scale deployment on Tencent WeChat Channels (~10M daily samples) further demonstrates its effectiveness, significantly mitigating AUC degradation, LogLoss increase, and prediction bias compared to standard distillation baselines.

cs.AI↗

arXiv2025

Practical Channel Estimation for Pinching-Antenna Systems: Serial vs. Parallel and Downlink vs. Uplink?

The practical channel estimation for pinching-antenna networks is investigated, in which an electromagnetic-compliant in-waveguide transmission model is exhibited, incorporating bidirectional power splitting, cumulative power leakage, and waveguide attenuation. Based on this model, the paper investigates two antenna activation protocols for channel estimation: a serial protocol based on one-by-one antenna activation and a parallel protocol utilizing a binary S-Matrix activation. The serial protocol is characterized by its superior numerical stability but a lack of array gain, whereas the parallel protocol theoretically offers array gain but suffers from severe performance degradation due to structural crosstalk from the non-orthogonal S-Matrix and ill-conditioning from cumulative leakage. Furthermore, the paper analyzes the fundamental commonalities and asymmetries between uplink and downlink channel estimation in pinching-antenna systems. Numerical results demonstrate that 1) in an ideal lossless model, the parallel protocol is superior to the serial protocol due to the array gain from simultaneous energy collection in uplink transmission; 2) in a practical model with physical losses, the serial protocol outperforms the parallel protocol, as the performance of the parallel protocol is degraded by the numerical instability from cumulative leakage, which outweighs the benefit of array gain; 3) For downlink channel estimation, the serial protocol is more suitable because its strategy of concentrating the entire power budget on one measurement, while the parallel protocol is more suitable for the uplink as it can make full use of array gain.

cs.IT↗

arXiv2025

Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval

Recent progress in text-video retrieval has been largely driven by contrastive learning. However, existing methods often overlook the effect of the modality gap, which causes anchor representations to undergo in-place optimization (i.e., optimization tension) that limits their alignment capacity. Moreover, noisy hard negatives further distort the semantics of anchors. To address these issues, we propose GARE, a Gap-Aware Retrieval framework that introduces a learnable, pair-specific increment $Δ_{ij}$ between text $t_i$ and video $v_j$, redistributing gradients to relieve optimization tension and absorb noise. We derive $Δ_{ij}$ via a multivariate first-order Taylor expansion of the InfoNCE loss under a trust-region constraint, showing that it guides updates along locally consistent descent directions. A lightweight neural module conditioned on the semantic gap couples increments across batches for structure-aware correction. Furthermore, we regularize $Δ$ through a variational information bottleneck with relaxed compression, enhancing stability and semantic consistency. Experiments on four benchmarks demonstrate that GARE consistently improves alignment accuracy and robustness, validating the effectiveness of gap-aware tension mitigation. Code is available at https://github.com/musicman217/GARE-text-video-retrieval.

cs.CV↗

arXiv2025

Frequency-Selective Modeling and Analysis for OFDM-Integrated Wideband Pinching-Antenna Systems

This letter investigates the integration of pinching-antenna systems (PASS) with orthogonal frequency division multiplexing (OFDM) to ensure their compatibility and to explore the frequency-selective behavior inherent to PASS. First, an end-to-end channel model for OFDM PASS is proposed based on electromagnetic-compliant modeling of waveguides and coupled-mode theory, which includes frequency-dependent waveguide attenuation, dispersion and antenna coupling effect. Furthermore, a critical dependence of the OFDM cyclic prefix (CP) overhead on the proximity of the operating frequency to the waveguide cutoff is revealed. Moreover, the phase misalignment effect across subcarriers in OFDM PASS is derived for an approximate pinching antenna location strategy based on path loss minimization, which reveals the phase misalignment is exacerbated for wider bandwidths and larger array size. Numerical results show that: 1) frequency-selective effects in OFDM PASS lead to substantial variations in subcarrier achievable rates, highlighting the necessity of operating above the waveguide cutoff frequency for effective communications; 2) waveguide dispersion mandates considerable CP overhead when operating near the cutoff frequency, severely impacting the spectral efficiency of OFDM PASS; and 3) the gentle linear waveguide attenuation in a practical PASS significantly more advantageous than the severe logarithmic path loss characteristic of fixed-location antennas.

cs.IT↗