arXiv subjects

Xueyi Wang

Publications and source records attributed to Xueyi Wang.

11 recordsLinked to original sources

arXiv2026

AdaptStress: Online Adaptive Learning for Interpretable and Personalized Stress Prediction Using Multivariate and Sparse Physiological Signals

Continuous stress forecasting could potentially contribute to lifestyle interventions. This paper presents a novel, explainable, and individualized approach for stress prediction using physiological data from consumer-grade smartwatches. We develop a time series forecasting model that leverages multivariate features, including heart rate variability, activity patterns, and sleep metrics, to predict stress levels across 16 temporal horizons (History window: 3, 5, 7, 9 days; forecasting window: 1, 3, 5, 7 days). Our evaluation involves 16 participants monitored for 10-15 weeks. We evaluate our approach across 16 participants, comparing against state-of-the-art time series models (Informer, TimesNet, PatchTST) and traditional baselines (CNN, LSTM, CNN-LSTM) across multiple temporal horizons. Our model achieved performance with an MSE of 0.053, MAE of 0.190, and RMSE of 0.226 in optimal settings (5-day input, 1-day prediction). A comparison with the baseline models shows that our model outperforms TimesNet, PatchTST, CNN-LSTM, LSTM, and CNN under all conditions, representing improvements of 36.9%, 25.5%, and 21.5% over the best baseline. According to the explanability analysis, sleep metrics are the most dominant and consistent stress predictors (importance: 1.1, consistency: 0.9-1.0), while activity features exhibit high inter-participant variability (0.1-0.2). Most notably, the model captures individual-specific patterns where identical features can have opposing effects across users, validating its personalization capabilities. These findings establish that consumer wearables, combined with adaptive and interpretable deep learning, can deliver relevant stress assessment adapted to individual physiological responses, providing a foundation for scalable, continuous, explainable mental health monitoring in real-world settings.

cs.LG↗

arXiv2025

Individualized and Interpretable Sleep Forecasting via a Two-Stage Adaptive Spatial-Temporal Model

Sleep quality impacts well-being. Therefore, healthcare providers and individuals need accessible and reliable forecasting tools for preventive interventions. This paper introduces an interpretable, individualized adaptive spatial-temporal model for predicting sleep quality. We designed a hierarchical architecture, consisting of parallel 1D convolutions with varying kernel sizes and dilated convolution, which extracts multi-resolution temporal patterns-short kernels capture rapid physiological changes, while larger kernels and dilation model slower trends. The extracted features are then refined through channel attention, which learns to emphasize the most predictive variables for each individual, followed by bidirectional LSTM and self-attention that jointly model both local sequential dynamics and global temporal dependencies. Finally, a two-stage adaptation strategy ensures the learned representations transfer effectively to new users. We conducted various experiments with five input window sizes (3, 5, 7, 9, and 11 days) and five prediction window sizes (1, 3, 5, 7, and 9 days). Our model consistently outperformed time series forecasting baseline approaches, including LSTM, Informer, PatchTST, and TimesNet. The best performance was achieved with a three-day input window and a one-day prediction window, yielding an RMSE of 0.216. Furthermore, the model demonstrated good predictive performance even for longer forecasting horizons (e.g., with a 0.257 RMSE for a three-day prediction window), highlighting its practical utility for real-world applications. We also conducted an explainability analysis to examine how different features influence sleep quality. These findings proved that the proposed framework offers a robust, adaptive, and explainable solution for personalized sleep forecasting using sparse data from commercial wearable devices.

cs.LG↗

arXiv2025

Personalized Sleep Prediction via Deep Adaptive Spatiotemporal Modeling and Sparse Data

A sleep forecast allows individuals and healthcare providers to anticipate and proactively address factors influencing restful rest, ultimately improving mental and physical well-being. This work presents an adaptive spatial and temporal model (AdaST-Sleep) for predicting sleep scores. Our proposed model combines convolutional layers to capture spatial feature interactions between multiple features and recurrent neural network layers to handle longer-term temporal health-related data. A domain classifier is further integrated to generalize across different subjects. We conducted several experiments using five input window sizes (3, 5, 7, 9, 11 days) and five predicting window sizes (1, 3, 5, 7, 9 days). Our approach consistently outperformed four baseline models, achieving its lowest RMSE (0.282) with a seven-day input window and a one-day predicting window. Moreover, the method maintained strong performance even when forecasting multiple days into the future, demonstrating its versatility for real-world applications. Visual comparisons reveal that the model accurately tracks both the overall sleep score level and daily fluctuations. These findings prove that the proposed framework provides a robust and adaptable solution for personalized sleep forecasting using sparse data from commercial wearable devices and domain adaptation techniques.

eess.SP↗

arXiv2025

OmniEraser: Remove Objects and Their Effects in Images with Paired Video-Frame Data

Inpainting algorithms have achieved remarkable progress in removing objects from images, yet still face two challenges: 1) struggle to handle the object's visual effects such as shadow and reflection; 2) easily generate shape-like artifacts and unintended content. In this paper, we propose Video4Removal, a large-scale dataset comprising over 100,000 high-quality samples with realistic object shadows and reflections. By constructing object-background pairs from video frames with off-the-shelf vision models, the labor costs of data acquisition can be significantly reduced. To avoid generating shape-like artifacts and unintended content, we propose Object-Background Guidance, an elaborated paradigm that takes both the foreground object and background images. It can guide the diffusion process to harness richer contextual information. Based on the above two designs, we present OmniEraser, a novel method that seamlessly removes objects and their visual effects using only object masks as input. Extensive experiments show that OmniEraser significantly outperforms previous methods, particularly in complex in-the-wild scenes. And it also exhibits a strong generalization ability in anime-style images. Datasets, models, and codes will be published.

cs.CV↗

arXiv2024

Enhanced Nonlinear Frequency Conversion Bandwidth through Birefringence induced Mode Hybridization

On-chip quantum information network requires qubit transfer between different wavelengths while preserving quantum coherence and entanglement, which needs broadband up-conversion available. Herein, we demonstrate a mode-hybridization based broadband nonlinear frequency conversion on X-cut thin film lithium niobate. With the spontaneous quasi-phase matching and quasi groupvelocity matching being simultaneously satisfied, broadband second harmonic generation with a 3-dB bandwidth up to 13 nm has been achieved in a micro-racetrack resonator. The same mechanism can work on the frequency conversion of the ultra-short pulse in the bent waveguide structure. This work will be beneficial to on-chip tunable frequency conversion and quantum light source generation on integrated photonic platforms, and further enable on-chip large-capacity multiplexing, multichannel optical information processing, and large quantum information networks.

physics.optics↗

arXiv2024

From pre-Lie bialgebras to phase spaces of Lie algebras: a categorical correspondence

This paper establishes a categorical framework for phase spaces of Lie algebras, pre-Lie bialgebras, Manin triples, classical s-matrices, and relative Rota-Baxter operators by introducing the concept of coherent homomorphisms. Starting with endo pre-Lie algebras (pre-Lie algebras equipped with endomorphisms), we extend classical constructions to this enhanced setting, which leads to the notion of coherent endomorphisms for each class of structures. Through polarization, these endomorphisms naturally generalize to coherent homomorphisms, establishing well-defined categories of these algebraic objects. Furthermore, mappings between categories are elevated to functors or equivalences, formalizing interconnections among the constructions. Finally, exploiting the categorical correspondence between s-matrices and relative Rota-Baxter operators, we develop cohomology and deformation of s-matrices, thereby bridging algebraic structures with category-theoretic methods.

math.RA↗

arXiv2023

Self-Amplification-Assisted Highly Efficient Integrated Laser

Light source is indispensable component in on-chip system. Compared with hybrid or heterogeneous integrated laser, monolithically integrated laser is more suitable for high density photonic integrated circuit (PIC) since the capability of large-scale manufacturing, lower active-passive coupling loss and less test complexity. Recent years have seen the spark of researches on rare-earth ion doped thin film lithium niobate (REI:TFLN), demonstrations have been made both in classical and quantum chips. However, low output power and limited quantum emitting efficiency hinder the application of the chip-scale laser source based on REI:TFLN. Here a highly efficient integrated laser assisted by cascaded amplifiers is proposed and experimentally prepared on Erbium-doped TFLN. A slope efficiency of 0.43% and a linewidth of 47.86 kHz are obtained. The maximum integrated laser power is 7.989 μW. Our results show a viable solution to improve efficiency by self-amplification without changing the intrinsic quantum emitting efficiency of the material, and our design has potential application in incorporating with functional devices such as optical communications, integrated quantum memory and quantum emission.

physics.optics↗

arXiv2023

EGOFALLS: A visual-audio dataset and benchmark for fall detection using egocentric cameras

Falls are significant and often fatal for vulnerable populations such as the elderly. Previous works have addressed the detection of falls by relying on data capture by a single sensor, images or accelerometers. In this work, we rely on multimodal descriptors extracted from videos captured by egocentric cameras. Our proposed method includes a late decision fusion layer that builds on top of the extracted descriptors. Furthermore, we collect a new dataset on which we assess our proposed approach. We believe this is the first public dataset of its kind. The dataset comprises 10,948 video samples by 14 subjects. We conducted ablation experiments to assess the performance of individual feature extractors, fusion of visual information, and fusion of both visual and audio information. Moreover, we experimented with internal and external cross-validation. Our results demonstrate that the fusion of audio and visual information through late decision fusion improves detection performance, making it a promising tool for fall prevention and mitigation.

cs.CV↗

arXiv2023

Enhanced temperature sensing by multi-mode coupling in an on-chip microcavity system

The micro-cavity is a promising sensor platform, any perturbation would disturb its linewidth, cause resonance shift or splitting. However, such sensing resolution is limited by the cavity's optical quality factor and mode volume. Here we propose and demonstrate in an on-chip integrated microcavity system that resolution of a self referenced sensor could be enhanced with multi mode coupling.

physics.optics↗

arXiv2019

Expansion of Solar Coronal Hot Electrons in an Inhomogeneous Magnetic Field: 1-D PIC Simulation

The expansion of hot electrons in flaring magnetic loops is crucial to understanding the dynamics of solar flares. In this paper we investigate, for the first time, the transport of hot electrons in a magnetic mirror field based on a 1-D particle-in-cell (PIC) simulation. The hot electrons with small pitch angle transport into the cold plasma, which leads to the generation of Langmuir waves in the cold plasma and ion acoustic waves in the hot plasma. The large pitch angle electrons can be confined by the magnetic mirror, resulting in the different evolution time scale between electron parallel and perpendicular temperature. This will cause the formation of electron temperature anisotropy, which can generate the whistler waves near the interface between hot electrons and cold electrons. The whistler waves can scatter the large pitch angle electrons to smaller value through the cyclotron resonance, leading to electrons escaping from the hot region. These results indicate that the whistler waves may play an important role in the transport of electrons in flaring magnetic loops. The findings from this study provide some new insights to understand the electron dynamics of solar flares.

physics.space-ph↗

arXiv2019

VAE-based Domain Adaptation for Speaker Verification

Deep speaker embedding has achieved satisfactory performance in speaker verification. By enforcing the neural model to discriminate the speakers in the training set, deep speaker embedding (called `x-vectors`) can be derived from the hidden layers. Despite its good performance, the present embedding model is highly domain sensitive, which means that it often works well in domains whose acoustic condition matches that of the training data (in-domain), but degrades in mismatched domains (out-of-domain). In this paper, we present a domain adaptation approach based on Variational Auto-Encoder (VAE). This model transforms x-vectors to a regularized latent space; within this latent space, a small amount of data from the target domain is sufficient to accomplish the adaptation. Our experiments demonstrated that by this VAE-adaptation approach, speaker embeddings can be easily transformed to the target domain, leading to noticeable performance improvement.

eess.AS↗