SearcharxivSearch

arXiv subjects

Xinyao Yang

Publications and source records attributed to Xinyao Yang.

5 recordsLinked to original sources

FusionGen: Feature Fusion-Based Few-Shot EEG Data Generation

Brain-computer interfaces (BCIs) provide potential for applications ranging from medical rehabilitation to cognitive state assessment by establishing direct communication pathways between the brain and external devices via electroencephalography (EEG). However, EEG-based BCIs are severely constrained by data scarcity and significant inter-subject variability, which hinder the generalization and applicability of EEG decoding models in practical settings. To address these challenges, we propose FusionGen, a novel EEG data generation framework based on disentangled representation learning and feature fusion. By integrating features across trials through a feature matching fusion module and combining them with a lightweight feature extraction and reconstruction pipeline, FusionGen ensures both data diversity and trainability under limited data constraints. Extensive experiments on multiple publicly available EEG datasets demonstrate that FusionGen significantly outperforms existing augmentation techniques, yielding notable improvements in classification accuracy.

cs.LG

The 1st International Workshop on Disentangled Representation Learning for Controllable Generation (DRL4Real): Methods and Results

This paper reviews the 1st International Workshop on Disentangled Representation Learning for Controllable Generation (DRL4Real), held in conjunction with ICCV 2025. The workshop aimed to bridge the gap between the theoretical promise of Disentangled Representation Learning (DRL) and its application in realistic scenarios, moving beyond synthetic benchmarks. DRL4Real focused on evaluating DRL methods in practical applications such as controllable generation, exploring advancements in model robustness, interpretability, and generalization. The workshop accepted 9 papers covering a broad range of topics, including the integration of novel inductive biases (e.g., language), the application of diffusion models to DRL, 3D-aware disentanglement, and the expansion of DRL into specialized domains like autonomous driving and EEG analysis. This summary details the workshop's objectives, the themes of the accepted papers, and provides an overview of the methodologies proposed by the authors.

cs.LG

Stability of the steady states in multidimensional reaction diffusion systems arising in combustion theory

We prove that the steady state of a class of multidimensional reaction-diffusion systems is asymptotically stable at the intersection of unweighted space and exponentially weighted Sobolev spaces, and pay particular attention to a special case, namely, systems of equations that arise in combustion theory. The steady-state solutions considered here are the end states of the traveling fronts associated with the systems, and thus the present results complement recent papers \cite{GLS1, GLS2, GLS3, GLSR, GLY} that study the stability of traveling fronts.

math.AP

Stability of a planar front in a multidimensional reaction-diffusion system

We study the planar front solution for a class of reaction diffusion equations in multidimensional space in the case when the essential spectrum of the linearization in the direction of the front touches the imaginary axis. At the linear level, the spectrum is stabilized by using an exponential weight. A-priori estimates for the nonlinear terms of the equation governing the evolution of the perturbations of the front are obtained when perturbations belong to the intersection of the exponentially weighted space with the original space without a weight. These estimates are then used to show that in the original norm, initially small perturbations to the front remain bounded, while in the exponentially weighted norm, they algebraically decay in time.

math.AP

Stable foliations near a traveling front for reaction diffusion systems

We establish the existence of a stable foliation in the vicinity of a traveling front solution for systems of reaction diffusion equations in one space dimension that arise in the study of chemical reactions models and solid fuel combustion. In this way we complement the orbital stability results from earlier papers by A. Ghazaryan, S. Schecter and Y. Latushkin. The essential spectrum of the differential operator obtained by linearization at the front touches the imaginary axis. In spaces with exponential weights, one can shift the spectrum to the left. We study the nonlinear equation on the intersection of the unweighted and weighted spaces. Small translations of the front form a center unstable manifold. For each small translation we prove the existence of a stable manifold containing the translated front and show that the stable manifolds foliate a small ball centered at the front.

math.DS