arXiv subjects

Xinxin Xu

Publications and source records attributed to Xinxin Xu.

9 recordsLinked to original sources

arXiv2026

SCALMU: Synthetically-trained Coupling of Adaptive Learned Multiplicative Updates for Hyperspectral-Multispectral Fusion

HyperSpectral-MultiSpectral Image (HSI-MSI) fusion aims to recover a high-resolution hyperspectral image from a low-resolution HSI and a high-resolution MSI. Classical methods such as Coupled Nonnegative Matrix Factorization (CNMF) benefit from a strong physical interpretability but suffer from inferior results compared to their deep-learning counterparts. To address this limitation, we propose SCALMU (Synthetically-trained Coupling of Adaptive Learned Multiplicative Updates), a novel blind unrolled neural network architecture that integrates adaptive learnable matrices within the classical framework of CNMF multiplicative updates, improving its results. Due to its architectural proximity with CNMF, the resulting algorithm preserves physical interpretability and nonnegativity constraints. To overcome the scarcity of supervised training data, we generate a synthetic HSI-MSI dataset using the dead leaves model and train SCALMU end-to-end under synthetic supervision. Experiments on several datasets show that SCALMU outperforms state-of-the-art methods and highlights the potential of blind fusion trained with synthetic data. The code is available at https://github.com/xinxinxu99/SCALMU.git

eess.IV↗

arXiv2026

Super-r\'esolution non supervis\'ee d'images hyperspectrales de t\'el\'ed\'etection utilisant un entra\^inement enti\`erement synth\'etique

Hyperspectral single image super-resolution (SISR) aims to enhance spatial resolution while preserving the rich spectral information of hyperspectral images. Most existing methods rely on supervised learning with high-resolution ground truth data, which is often unavailable in practice. To overcome this limitation, we propose an unsupervised learning approach based on synthetic abundance data. The hyperspectral image is first decomposed into endmembers and abundance maps through hyperspectral unmixing. A neural network is then trained to super-resolve these maps using data generated with the dead leaves model, which replicates the statistical properties of real abundances. The final super-resolution hyperspectral image is reconstructed by recombining the super-resolved abundance maps with the endmembers. Experimental results demonstrate the effectiveness of our method and the relevance of synthetic data for training.

eess.IV↗

arXiv2026

Synthetic Abundance Maps for Unsupervised Super-Resolution of Hyperspectral Remote Sensing Images

Hyperspectral single image super-resolution (HS-SISR) aims to enhance the spatial resolution of hyperspectral images to fully exploit their spectral information. While considerable progress has been made in this field, most existing methods are supervised and require ground truth data for training-data that is often unavailable in practice. To overcome this limitation, we propose a novel unsupervised training framework for HS-SISR, based on synthetic abundance data, where no high-resolution ground-truth reference is required for training. The approach begins by unmixing the hyperspectral image into endmembers and abundances. A neural network is then trained to perform abundance super-resolution using synthetic abundances only. These synthetic abundance maps are generated from a dead leaves model whose characteristics are inherited from the low-resolution image to be super-resolved and from the known point spread function (PSF) of the hyperspectral sensor. This trained network is subsequently used to enhance the spatial resolution of the original image's abundances, and the final super-resolution hyperspectral image is reconstructed by combining them with the endmembers. Experimental results demonstrate both the training value of the synthetic data and the effectiveness of the proposed method across 3 datasets, 3 scaling factors, and several evaluation metrics. The code is available at https://github.com/xinxinxu99/SISR-DL.git

eess.IV↗

arXiv2026

Unsupervised Super-Resolution of Hyperspectral Remote Sensing Images Using Fully Synthetic Training

Considerable work has been dedicated to hyperspectral single image super-resolution to improve the spatial resolution of hyperspectral images and fully exploit their potential. However, most of these methods are supervised and require some data with ground truth for training, which is often non-available. To overcome this problem, we propose a new unsupervised training strategy for the super-resolution of hyperspectral remote sensing images, based on the use of synthetic abundance data. Its first step decomposes the hyperspectral image into abundances and endmembers by unmixing. Then, an abundance super-resolution neural network is trained using synthetic abundances, which are generated using the dead leaves model in such a way as to faithfully mimic real abundance statistics. Next, the spatial resolution of the considered hyperspectral image abundances is increased using this trained network, and the high resolution hyperspectral image is finally obtained by recombination with the endmembers. Experimental results show the training potential of the synthetic images, and demonstrate the method effectiveness.

eess.IV↗

arXiv2025

Two Heads are Better than One: Distilling Large Language Model Features Into Small Models with Feature Decomposition and Mixture

Market making (MM) through Reinforcement Learning (RL) has attracted significant attention in financial trading. With the development of Large Language Models (LLMs), more and more attempts are being made to apply LLMs to financial areas. A simple, direct application of LLM as an agent shows significant performance. Such methods are hindered by their slow inference speed, while most of the current research has not studied LLM distillation for this specific task. To address this, we first propose the normalized fluorescent probe to study the mechanism of the LLM's feature. Based on the observation found by our investigation, we propose Cooperative Market Making (CMM), a novel framework that decouples LLM features across three orthogonal dimensions: layer, task, and data. Various student models collaboratively learn simple LLM features along with different dimensions, with each model responsible for a distinct feature to achieve knowledge distillation. Furthermore, CMM introduces an H\'{a}jek-MoE to integrate the output of the student models by investigating the contribution of different models in a kernel function-generated common feature space. Extensive experimental results on four real-world market datasets demonstrate the superiority of CMM over the current distillation method and RL-based market-making strategies.

cs.AI↗

arXiv2019

Production of the exotic neutron-deficient isotopes near N, Z = 50 in multinucleon transfer reactions

The multinucleon transfer reaction in the collisions of $^{40}$Ca+$^{124}$Sn at $E_{\textrm{c.m.}}=128.5$ MeV is investigated by using the improved quantum molecular dynamics model. The measured angular distributions and isotopic distributions of the products are reproduced reasonably well by the calculations. The multinucleon transfer reactions of $^{40}$Ca+$^{112}$Sn, $^{58}$Ni+$^{112}$Sn, $^{106}$Cd+$^{112}$Sn, and $^{48}$Ca+$^{112}$Sn are also studied. It shows that the combinations of neutron-deficient projectile and target are advantageous to produce the exotic neutron-deficient nuclei near $N, Z$ = 50. The charged particles emission plays an important role at small impact parameters in the deexcitation processes of the system. The production cross sections of the exotic neutron-deficient nuclei in multinucleon transfer reactions are much larger than those measured in the fragmentation and fusion-evaporation reactions. Several new neutron-deficient nuclei can be produced in $^{106}$Cd+$^{112}$Sn reaction. The corresponding production cross sections for the new neutron-deficient nuclei, $^{101,102}$Sb, $^{103}$Te, and $^{106,107}$I, are 2.0 nb, 4.1 nb, 6.5 nb, 0.4 $μ$b and 1.0 $μ$b, respectively.

nucl-th↗

arXiv2018

Production of new neutron-rich heavy nuclei with $Z=56-80$ in the multinucleon transfer reactions of $^{136}$Xe+$^{198}$Pt

The multinucleon transfer reactions in collisions of $^{136}$Xe+$^{198}$Pt at incident energies $E_{\textrm{lab}}=$5.25, 6.20, 7.98, 10.0, and 15.0 MeV/nucleon are investigated by using the improved quantum molecular dynamics model. It is found that 6.20 MeV/nucleon is the optimal incident energy for producing the neutron-rich heavy nuclei. About 80 unknown neutron-rich nuclei might be produced in this reaction with cross sections from 10$^{-6}$ to 10$^{-2}$ mb. The angular distributions of the neutron-rich isotopes are predicted.

nucl-th↗

arXiv2018

Isospin equilibration in multinucleon transfer reaction at near-barrier energies

The isospin equilibration process in multinucleon transfer reaction is investigated by using the improved quantum molecular dynamics model. The collision processes of $^{124}$Xe+$^{208}$Pb at near-barrier energy are studied with different symmetry energy coefficients. We find that neutrons transfer happens earlier than protons. The large symmetry energy promote the transfer of neutrons. The neutron flow from the target to projectile is along the low-density path of neck. The isospin equilibration process in $^{58}$Ni+$^{208}$Pb reaction is also investigated and compared with available experimental data. It shows that $N/Z$ values of the projectile-like products increase rapidly with increasing mass transfer. The complete isospin equilibration events are located in the region of $120<A<150$ which are produced in symmetric fission-type reactions.

nucl-th↗

arXiv2016

HISTORY: An Efficient and Robust Algorithm for Noisy 1-bit Compressed Sensing

We consider the problem of sparse signal recovery from 1-bit measurements. Due to the noise present in the acquisition and transmission process, some quantized bits may be flipped to their opposite states. These sign flips may result in severe performance degradation. In this study, a novel algorithm, termed HISTORY, is proposed. It consists of Hamming support detection and coefficients recovery. The HISTORY algorithm has high recovery accuracy and is robust to strong measurement noise. Numerical results are provided to demonstrate the effectiveness and superiority of the proposed algorithm.

cs.IT↗