Beijing Jiaotong University

Information-Bottlenecked Variational Autoencoder for Top- N Recommendation via Gating Mechanism

Generative models for top- \(N\) recommendation have garnered significant attention, with Variational Autoencoder (VAE) emerging as a promising approach for modeling user preferences. Yet, traditional VAE-based models encounter two major challenges: simplistic priors may cause po...

Scaling of Urban 3D Form with Emissions and GDP in Chinese Cities

Abstract Rapid urbanization has permanently transformed the face of cities. As the physical manifestation of these transformations, urban form provides a key basis for examining its connections with urban outcomes. While previous studies have primarily focused on two-dimensional...

Experimental and simulation study on the micro-explosion characteristics of ethanol/biodiesel droplets

Biodiesel exhibits superior properties such as high cetane number, high flash point, and high lubricity, which enhance its ignition, storage, and usage characteristics. However, its high density and viscosity can impede fuel atomization. This issue can be effectively mitigated by...

Understanding Influencing Mechanisms of Subway Station-Line Peak Deviation Time: Spatial Econometric Perspective

The peak deviation time (PDT) between subway stations and lines is a critical temporal misalignment in urban rail systems, yet it remains understudied. Accurately understanding PDT is essential for calibrating station-level peak-hour time (PHT) during planning, which directly inf...

Large maximal subgroups of almost simple classical groups

Abstract A proper subgroup 𝐻 of a finite group 𝐺 is called large if | H | 3 ≥ | G | \lvert H\rvert^{3}\geq\lvert G\rvert . This paper characterises all the large maximal subgroups of almost simple groups whose socle is a finite simple classical group. For such a socle G 0 G_{0} a...

Research and Application of Water Load and Internal Forces in Composite Linings of Subsea Tunnels

Water loads on subsea tunnel linings are governed by the combined effects of seepage, structural deformation, and the grouting zone. However, these factors are often treated separately in existing analytical studies. In this study, an analytical model is developed for steady-stat...

Dynamic User Equilibrium for Electric Vehicle Departure Time and Path–Charging Choices with Wireless and Fast Charging Services

This study investigates how coordinated wireless and fast charging services reshape electric vehicle departure time and path–charging choices when a trip-level charging requirement must be completed before arrival. A multi-class dynamic user equilibrium model is formulated for ro...

STOD: Sparse Tensor Train Optimization via Orthogonal Decomposition for High-Dimensional Learning

This paper proposes a novel Tensor Train (TT)-based tensor-on-tensor regression optimization framework for variable selection based on mode-1 hyperslice sparsity. The formulation incorporates an l2,0-regularized term on the first TT-core while imposing Stiefel manifold constraint...

MDiffFR: Modality-Guided Diffusion Generation for Cold-start Items in Federated Recommendation

Federated recommendations (FRs) provide personalized services while preserving user privacy by keeping user data on local clients, which has attracted significant attention in recent years. Due to the strict privacy constraints inherent in FRs, access to user-item interaction dat...

Investigating the impact of visual reference grids on the calibration of auditory distance perception from different directions in VR

Distance perception plays a critical role in shaping user experience in Virtual Reality (VR) environments. While auditory distance perception has been extensively studied in virtual acoustics, much less is known about how structured visual reference grids and calibration procedur...

Trust-Aware Enterprise Credit Risk Prediction via Variational Autoencoder in Supply Chain Finance

The growing complexity of transaction networks, dynamic enterprise interactions, and rising financial uncertainties in modern supply chains have made accurate credit risk assessment for small and medium-sized enterprises (SMEs) in supply chain finance (SCF) a critical priority. H...

The Impact of Patient Capital on Innovation Quantity and Quality Among SMEs

Drawing on panel data from firms listed on the SME Board and Growth Enterprise Market (GEM) between 2010 and 2024, this study examines how patient capital influences SME innovation. It considers both the quantity and quality of innovation and investigates the underlying mechanism...

Synergistic Dual‐Channel Charge Separation in Cr‐TiO 2‐x for Solar H 2 O 2 Synthesis

ABSTRACT Solar‐driven hydrogen peroxide (H 2 O 2 ) synthesis offers a sustainable alternative to the energy‐intensive anthraquinone process, yet its efficiency is fundamentally constrained by the inability to simultaneously harness photogenerated electrons and holes in a balanced...

Synergistic Dual‐Channel Charge Separation in Cr‐TiO 2‐x for Solar H 2 O 2 Synthesis

ABSTRACT Solar‐driven hydrogen peroxide (H 2 O 2 ) synthesis offers a sustainable alternative to the energy‐intensive anthraquinone process, yet its efficiency is fundamentally constrained by the inability to simultaneously harness photogenerated electrons and holes in a balanced...

On the Convergence of JPEG Multiple Compression for Digital Forensics

Convergence means that the limit of the sequence will tend to a certain value. In some forensic scenarios, if the same manipulation is repeatedly performed on an image, the resulting image sequence tends to converge. Credible forensics can be achieved by utilizing this convergenc...

A Quality-Aware Multimodal Reliability Framework for Health Assessment and Remaining Useful Life Prediction of Cold-Region Tunnels

This study proposes a quality-aware multimodal framework for health-state assessment and remaining useful life (RUL) prediction of cold-region tunnels. The framework integrates structural-response, environmental, apparent-defect, and engineering-inspectiondata, with the apparent-...

Molecular Doping Induced High‐Performance Filterless Red‐Light Organic/Perovskite Photodetectors for Quantitative Analysis of Medical Photosensitizer Materials

ABSTRACT Red‐light narrowband photodetectors play indispensable roles in the photomedical industry, particularly in applications such as fluorescence imaging, photodynamic therapy, and non‐invasive optical diagnostics. Herein, we introduce a novel device architecture (OP‐PDs) tha...

Single-frame vehicle trajectory prediction via neural ODE-based motion-state forecasting

Accurate short-horizon vehicle trajectory prediction is critical for autonomous driving and vehicle-to-everything (V2X) communication. Most existing deep learning-based prediction methods rely on fixed-length historical trajectories and multi-agent context, which may be unavailab...

Ultra-low frequency Vibration Isolation Based on a Symmetric-Converging Series Horizontal-Spring Quasi-Zero-Stiffness Mechanism

Low-frequency vibrations, especially those in the ultra-low frequency range below 1 Hz, are difficult to suppress due to their long wavelengths, slow attenuation, and strong penetration. To address this challenge, a novel quasi-zero-stiffness (QZS) isolation mechanism is proposed...

Risk Assessment of Post-Earthquake Gas Explosion Disaster Chains in High-Rise Residential Buildings: A Fuzzy Bayesian and Complex Network Approach

To address the challenges in risk prevention and control of post-earthquake gas explosions in high-rise buildings and the deficiencies of traditional methods in handling uncertainty, this paper conducts a risk evolution analysis from the perspectives of fuzzy Bayesian networks (F...