arXiv subjects

Yi Hong

Publications and source records attributed to Yi Hong.

At least 19 recordsLinked to original sources

arXiv2026

Sensing-Assisted Anti-Blockage Pinching-Antenna Systems For Indoor Immersive Communications

Indoor immersive communication, empowered by millimeter-wave (mmWave) technologies, is a key enabler for 6G-ready Extended Reality (XR) applications. However, mmWave signals are highly susceptible to dynamic obstacles (e.g., moving humans), leading to frequent line-of-sight (LoS) blockages. Moreover, although massive-antenna systems have been widely adopted indoors, their fixed-site deployment limits their effectiveness in wide and dynamic indoor environments such as factories and offices. To address these challenges, we present a sensing-assisted anti-blockage Pinching-Antenna Systems (PASS) solution for indoor immersive communications. By mechanically moving a pinching antenna (PA) along a dielectric waveguide, PASS can flexibly extend the service coverage and re-establish LoS links in wide indoor scenarios. Besides, our solution integrates sensing capabilities into PASS, where the PA executes controlled micro-movements to transmit and receive chirp radar signals, enabling real-time sensing of obstacles. Based on the sensing results, we design an anti-blockage PA movement strategy that detects blocked regions along the waveguide and proactively moves the PA to safe zones. Accordingly, the whole system performs a cycled "sensing-movement-communication" workflow with each cycle completed in a limited frame to ensure both timely sensing and communication. To ensure system efficiency, a closed-form pre-optimized solution of sensing parameters is derived under practical mechanical and sensing constraints. Extensive experiments show that our solution reduces blockage-induced outages to near-zero while sustaining the effective throughput required by indoor immersive applications.

eess.SP↗

arXiv2026

Channel Estimation for OTFS Systems With Overspread Doppler Shifts

In this paper, we consider an orthogonal time frequency space (OTFS) system in time-varying channels with overspread Doppler shifts, typically found in non-terrestrial multi-satellite links. The overspread Doppler shifts with magnitude greater than half of the subcarrier spacing, result in aliased Doppler shifts in the delay-Doppler (DD) domain due to the OTFS modulo operation. This makes channel estimation very challenging and the traditional channel estimation methods become ineffective. To address this challenge, we propose a DD training frame and a two-stage channel estimation method. The training frame comprises a cosine pilot signal and a pilot symbol. In the first stage of the channel estimation, the pilot symbol in the DD domain is utilized to estimate the delays, aliased Doppler shifts, and channel gains of the propagation paths. In the second stage, the received time domain signal is converted into the frequency domain to detect the peaks of all the Doppler shifts using the cosine pilot signal. Then, we present a threshold-based method to pair the estimated actual Doppler shifts with their corresponding delays and channel gains. The complexity of the proposed channel estimation is also discussed. Finally, the performance of the proposed channel estimation is validated in terms of the normalized mean square error (NMSE) and bit error rate (BER) in various scenarios.

cs.IT↗

arXiv2026

CARA: Concept-Aware Risk Attention for Interpretable Collision Anticipation

Collision anticipation in autonomous driving requires not only accurate early warnings but also interpretable reasoning about what risk factors are being tracked and how risk evolves over time. Existing methods fall short in this regard: feature-driven models are opaque, post-hoc explanations often lack fidelity, and concept-based methods are mostly designed for static recognition rather than dynamic driving scenes. We propose CARA (Concept-Aware Risk Attention), an intrinsically interpretable spatio-temporal framework for collision anticipation. CARA derives domain-grounded risk concepts from accident narratives, aligns them with video frames via vision-language similarity, and organizes them into evolving concept trajectories. These trajectories provide explicit risk evidence that guides spatial attention, temporal attention, and anticipation, allowing semantic concepts to directly influence both where the model attends and how it predicts risk over time. By treating semantic risk factors as dynamic intermediate evidence rather than auxiliary post-hoc explanations, CARA tightly couples interpretability with the predictive process. Extensive experiments on three benchmarks show that CARA consistently improves anticipation accuracy and warning earliness over strong baselines, while providing sparse and semantically grounded concept evidence.

cs.MM↗

arXiv2026

Corporate Bond Yield Curve Modeling: A Rating-Based Regime-Switching Generalized CIR Approach

Persistent shifts in term-structure dynamics undermine the stability of single-regime models in long samples. We develop an arbitrage-free regime-switching generalized CIR (RS-GCIR) model that jointly prices the Chinese government bond (CGB) curve and corporate bond curves. To capture the systematic transmission from interest-rate conditions to credit spreads, we structure the model into two blocks and price corporate bonds conditional on the prevailing rate regime. The rate block features a two-state RS-GCIR short-rate process estimated from CGB zero-coupon curves, while the credit block embeds CIR-type credit factors in an intensity-based framework for rating migration and default. We implement a block-recursive Unscented Kalman Filter (UKF) procedure--filtering the rate block first and the credit block next--using weekly data from 2014--2025, a period that begins with the onset of China's modern corporate default cycle. We identify two persistent rate regimes with distinct level--volatility profiles. Relative to single-regime benchmarks, regime switching improves joint curve fit, delivers economically interpretable filtered regime probabilities, and sharpens the decomposition of corporate yields into discounting and credit compensation.

q-fin.PR↗

arXiv2026

Sky-Ear: An Unmanned Aerial Vehicle-Enabled Victim Sound Detection and Localization System

Unmanned Aerial Vehicles (UAVs) are increasingly deployed in search-and-rescue (SAR) missions, yet continuous and reliable victim detection and localization remain challenging due to on-board hardware constraints. This paper designs an UAV-Enabled Victim Sound Detection and Localization System (called ``Sky-Ear'' for brevity) to achieve energy-efficient acoustic sensing and sound detection for SAR. Sky-Ear enables the ``ear'' of the UAV with a circular-shaped microphone array, and the array conducts continuous audio recordings during the UAV's flight. In Sky-Ear, a two-stage (Sentinel and Responder) audio processing method is developed for energy-consuming and highly reliable sound detection. In the Sentinel stage, a Masking autoencoder (MAE)-based sound detection mechanism is designed to analyze frequency-time acoustic features. For improved precision, a continuous localization method is designed by optimizing detected directions from multiple observations. Extensive simulation experiments are conducted to validate the system's performance in terms of victim detection accuracy and localization error.

eess.AS↗

arXiv2026

AD-Reasoning: Multimodal Guideline-Guided Reasoning for Alzheimer's Disease Diagnosis

Alzheimer's disease (AD) diagnosis requires integrating neuroimaging with heterogeneous clinical evidence and reasoning under established criteria, yet most multimodal models remain opaque and weakly guideline-aligned. We present AD-Reasoning, a multimodal framework that couples structural MRI with six clinical modalities and a rule-based verifier to generate structured, NIA-AA-consistent diagnoses. AD-Reasoning combines modality-specific encoders, bidirectional cross-attention fusion, and reinforcement fine-tuning with verifiable rewards that enforce output format, guideline evidence coverage, and reasoning--decision consistency. We also release AD-MultiSense, a 10,378-visit multimodal QA dataset with guideline-validated rationales built from ADNI/AIBL. On AD-MultiSense, AD-Reasoning achieves state-of-the-art diagnostic accuracy and produces structured rationales that improve transparency over recent baselines, while providing transparent rationales.

cs.CV↗

arXiv2026

EMAD: Evidence-Centric Grounded Multimodal Diagnosis for Alzheimer's Disease

Deep learning models for medical image analysis often act as black boxes, seldom aligning with clinical guidelines or explicitly linking decisions to supporting evidence. This is especially critical in Alzheimer's disease (AD), where predictions should be grounded in both anatomical and clinical findings. We present EMAD, a vision-language framework that generates structured AD diagnostic reports in which each claim is explicitly grounded in multimodal evidence. EMAD uses a hierarchical Sentence-Evidence-Anatomy (SEA) grounding mechanism: (i) sentence-to-evidence grounding links generated sentences to clinical evidence phrases, and (ii) evidence-to-anatomy grounding localizes corresponding structures on 3D brain MRI. To reduce dense annotation requirements, we propose GTX-Distill, which transfers grounding behavior from a teacher trained with limited supervision to a student operating on model-generated reports. We further introduce Executable-Rule GRPO, a reinforcement fine-tuning scheme with verifiable rewards that enforces clinical consistency, protocol adherence, and reasoning-diagnosis coherence. On the AD-MultiSense dataset, EMAD achieves state-of-the-art diagnostic accuracy and produces more transparent, anatomically faithful reports than existing methods. We will release code and grounding annotations to support future research in trustworthy medical vision-language models.

cs.CV↗

arXiv2025

Ultimate Forward Rate Prediction and its Application to Bond Yield Forecasting: A Machine Learning Perspective

This study focuses on forecasting the ultimate forward rate (UFR) and developing a UFRbased bond yield prediction model using data from Chinese treasury bonds and macroeconomic variables spanning from December 2009 to December 2024. The de Kort-Vellekooptype methodology is applied to estimate the UFR, incorporating the optimal turning parameter determination technique proposed in this study, which helps mitigate anomalous fluctuations. In addition, both linear and nonlinear machine learning techniques are employed to forecast the UFR and ultra-long-term bond yields. The results indicate that nonlinear machine learning models outperform their linear counterparts in forecasting accuracy. Incorporating macroeconomic variables, particularly price index-related variables, significantly improves the accuracy of predictions. Finally, a novel UFR-based bond yield forecasting model is developed, demonstrating superior performance across different bond maturities.

q-fin.ST↗

arXiv2025

Harmonizing Generalization and Specialization: Uncertainty-Informed Collaborative Learning for Semi-supervised Medical Image Segmentation

Vision foundation models have demonstrated strong generalization in medical image segmentation by leveraging large-scale, heterogeneous pretraining. However, they often struggle to generalize to specialized clinical tasks under limited annotations or rare pathological variations, due to a mismatch between general priors and task-specific requirements. To address this, we propose Uncertainty-informed Collaborative Learning (UnCoL), a dual-teacher framework that harmonizes generalization and specialization in semi-supervised medical image segmentation. Specifically, UnCoL distills both visual and semantic representations from a frozen foundation model to transfer general knowledge, while concurrently maintaining a progressively adapting teacher to capture fine-grained and task-specific representations. To balance guidance from both teachers, pseudo-label learning in UnCoL is adaptively regulated by predictive uncertainty, which selectively suppresses unreliable supervision and stabilizes learning in ambiguous regions. Experiments on diverse 2D and 3D segmentation benchmarks show that UnCoL consistently outperforms state-of-the-art semi-supervised methods and foundation model baselines. Moreover, our model delivers near fully supervised performance with markedly reduced annotation requirements.

cs.CV↗

arXiv2025

Ergodic Capacity and Optimal Handover in Satellite Mega-Constellations under Finite Serving Times

Existing analyses of ergodic capacity in satellite mega-constellations often rely on restrictive serving time assumptions or become intractable under realistic handover strategies. This paper develops a framework for characterising the ergodic capacity of low-Earth-orbit (LEO) mega-constellation links under arbitrary handover strategies and serving times. The user--satellite link is modelled as shadowed-Rician fading, and a semi-stochastic satellite channel with persistence is introduced in which visible satellites are drawn from a non-homogeneous binomial point process (NBPP) at each handover and the selected satellite is then propagated using circular orbit dynamics. Under uncoordinated handover decisions, this yields independent serving periods and enables a renewal-theoretic derivation of persistent capacity. This capacity is related to the non-persistent capacity from prior work, and closed-form bounds are provided for efficient evaluation. Optimal handover is then formulated as a non-linear fractional program, yielding an explicit decision rule via a variant of Dinkelbach's algorithm. The results show that a simpler strategy that maximises serving capacity closely approximates the optimum while performing best under SGP4-based orbit prediction and mega-constellation simulation.

cs.IT↗

arXiv2025

Enhancing 3D Medical Image Understanding with Pretraining Aided by 2D Multimodal Large Language Models

Understanding 3D medical image volumes is critical in the medical field, yet existing 3D medical convolution and transformer-based self-supervised learning (SSL) methods often lack deep semantic comprehension. Recent advancements in multimodal large language models (MLLMs) provide a promising approach to enhance image understanding through text descriptions. To leverage these 2D MLLMs for improved 3D medical image understanding, we propose Med3DInsight, a novel pretraining framework that integrates 3D image encoders with 2D MLLMs via a specially designed plane-slice-aware transformer module. Additionally, our model employs a partial optimal transport based alignment, demonstrating greater tolerance to noise introduced by potential noises in LLM-generated content. Med3DInsight introduces a new paradigm for scalable multimodal 3D medical representation learning without requiring human annotations. Extensive experiments demonstrate our state-of-the-art performance on two downstream tasks, i.e., segmentation and classification, across various public datasets with CT and MRI modalities, outperforming current SSL methods. Med3DInsight can be seamlessly integrated into existing 3D medical image understanding networks, potentially enhancing their performance. Our source code, generated datasets, and pre-trained models will be available at https://github.com/Qybc/Med3DInsight.

cs.CV↗

arXiv2025

Stochastic Channel Models for Satellite Mega-Constellations

A general satellite channel model is proposed for communications between a rapidly moving low Earth orbit (LEO) satellite in a mega-constellation and a stationary user on Earth. The channel uses a non-homogeneous binomial point process (NBPP) for modelling the satellite positions, marked with an ascending/descending binary random variable for modelling the satellite directions. Using the marked NBPP, we derive the probability distributions of power gain, propagation delay, and Doppler shift, resulting in a stochastic signal propagation model for the mega-constellation geometry in isolation of other effects. This forms the basis for our proposed channel model as a randomly time-varying channel. The scattering function of this channel is derived to characterise how the received power is spread in the delay-Doppler domain. Global channel parameters such as path loss and channel spread are analysed in terms of the scattering function. The channel statistics and the global channel parameters closely match realistic orbit simulations of the Starlink constellation.

cs.IT↗

arXiv2025

HoloDx: Knowledge- and Data-Driven Multimodal Diagnosis of Alzheimer's Disease

Accurate diagnosis of Alzheimer's disease (AD) requires effectively integrating multimodal data and clinical expertise. However, existing methods often struggle to fully utilize multimodal information and lack structured mechanisms to incorporate dynamic domain knowledge. To address these limitations, we propose HoloDx, a knowledge- and data-driven framework that enhances AD diagnosis by aligning domain knowledge with multimodal clinical data. HoloDx incorporates a knowledge injection module with a knowledge-aware gated cross-attention, allowing the model to dynamically integrate domain-specific insights from both large language models (LLMs) and clinical expertise. Also, a memory injection module with a designed prototypical memory attention enables the model to retain and retrieve subject-specific information, ensuring consistency in decision-making. By jointly leveraging these mechanisms, HoloDx enhances interpretability, improves robustness, and effectively aligns prior knowledge with current subject data. Evaluations on five AD datasets demonstrate that HoloDx outperforms state-of-the-art methods, achieving superior diagnostic accuracy and strong generalization across diverse cohorts. The source code will be released upon publication acceptance.

cs.CV↗

arXiv2024

An Event-centric Framework for Predicting Crime Hotspots with Flexible Time Intervals

Predicting crime hotspots in a city is a complex and critical task with significant societal implications. Numerous spatiotemporal correlations and irregularities pose substantial challenges to this endeavor. Existing methods commonly employ fixed-time granularities and sequence prediction models. However, determining appropriate time granularities is difficult, leading to inaccurate predictions for specific time windows. For example, users might ask: What are the crime hotspots during 12:00-20:00? To address this issue, we introduce FlexiCrime, a novel event-centric framework for predicting crime hotspots with flexible time intervals. FlexiCrime incorporates a continuous-time attention network to capture correlations between crime events, which learns crime context features, representing general crime patterns across time points and locations. Furthermore, we introduce a type-aware spatiotemporal point process that learns crime-evolving features, measuring the risk of specific crime types at a given time and location by considering the frequency of past crime events. The crime context and evolving features together allow us to predict whether an urban area is a crime hotspot given a future time interval. To evaluate FlexiCrime's effectiveness, we conducted experiments using real-world datasets from two cities, covering twelve crime types. The results show that our model outperforms baseline techniques in predicting crime hotspots over flexible time intervals.

cs.LG↗

arXiv2024

NODER: Image Sequence Regression Based on Neural Ordinary Differential Equations

Regression on medical image sequences can capture temporal image pattern changes and predict images at missing or future time points. However, existing geodesic regression methods limit their regression performance by a strong underlying assumption of linear dynamics, while diffusion-based methods have high computational costs and lack constraints to preserve image topology. In this paper, we propose an optimization-based new framework called NODER, which leverages neural ordinary differential equations to capture complex underlying dynamics and reduces its high computational cost of handling high-dimensional image volumes by introducing the latent space. We compare our NODER with two recent regression methods, and the experimental results on ADNI and ACDC datasets demonstrate that our method achieves the state-of-the-art performance in 3D image regression. Our model needs only a couple of images in a sequence for prediction, which is practical, especially for clinical situations where extremely limited image time series are available for analysis. Our source code is available at https://github.com/ZedKing12138/NODER-pytorch.

cs.CV↗

arXiv2024

OTFS Channel Estimation and Detection for Channels with Very Large Delay Spread

In low latency applications and in general, for overspread channels, channel delay spread is a large percentage of the transmission frame duration. In this paper, we consider OTFS in an overspread channel exhibiting a delay spread that exceeds the block duration in a frame, where traditional channel estimation (CE) fails. We propose a two-stage CE method based on a delay-Doppler (DD) training frame, consisting of a dual chirp converted from time domain and a higher power pilot. The first stage employs a DD domain embedded pilot CE to estimate the aliased delays (due to modulo operation) and Doppler shifts, followed by identifying all the underspread paths not coinciding with any overspread path. The second stage utilizes time domain dual chirp correlation to estimate the actual delays and Doppler shifts of the remaining paths. This stage also resolves ambiguity in estimating delays and Doppler shifts for paths sharing same aliased delay. Furthermore, we present a modified low-complexity maximum ratio combining (MRC) detection algorithm for OTFS in overspread channels. Finally, we evaluate performance of OTFS using the proposed CE and the modified MRC detection in terms of normalized mean square error (NMSE) and bit error rate (BER).

cs.IT↗

arXiv2024

Med3DInsight: Enhancing 3D Medical Image Understanding with 2D Multi-Modal Large Language Models

Understanding 3D medical image volumes is a critical task in the medical domain. However, existing 3D convolution and transformer-based methods have limited semantic understanding of an image volume and also need a large set of volumes for training. Recent advances in multi-modal large language models (MLLMs) provide a new and promising way to understand images with the help of text descriptions. However, most current MLLMs are designed for 2D natural images. To enhance the 3D medical image understanding with 2D MLLMs, we propose a novel pre-training framework called Med3DInsight, which marries existing 3D image encoders with 2D MLLMs and bridges them via a designed Plane-Slice-Aware Transformer (PSAT) module. Extensive experiments demonstrate our SOTA performance on two downstream segmentation and classification tasks, including three public datasets with CT and MRI modalities and comparison to more than ten baselines. Med3DInsight can be easily integrated into any current 3D medical image understanding network and improves its performance by a good margin.

cs.CV↗

arXiv2024

AliFuse: Aligning and Fusing Multi-modal Medical Data for Computer-Aided Diagnosis

Medical data collected for diagnostic decisions are typically multimodal, providing comprehensive information on a subject. While computer-aided diagnosis systems can benefit from multimodal inputs, effectively fusing such data remains a challenging task and a key focus in medical research. In this paper, we propose a transformer-based framework, called Alifuse, for aligning and fusing multimodal medical data. Specifically, we convert medical images and both unstructured and structured clinical records into vision and language tokens, employing intramodal and intermodal attention mechanisms to learn unified representations of all imaging and non-imaging data for classification. Additionally, we integrate restoration modeling with contrastive learning frameworks, jointly learning the high-level semantic alignment between images and texts and the low-level understanding of one modality with the help of another. We apply Alifuse to classify Alzheimer's disease, achieving state-of-the-art performance on five public datasets and outperforming eight baselines.

cs.CV↗