The goal of multispectral and hyperspectral image fusion (MHIF) is to generate high-quality images that simultaneously possess rich spectral information and fine spatial details. However, due to the inherent trade-off between spectral and spatial information and the limited availability of observations, this task is fundamentally ill-posed. Previous studies have not effectively addressed the ill-posed nature caused by data misalignment. To tackle this challenge, we propose a fusion framework named PIF-Net, which explicitly incorporates ill-posed priors to effectively fuse multispectral images and hyperspectral images. To balance global spectral modeling with computational efficiency, we design a method based on an invertible Mamba architecture that maintains information consistency during feature transformation and reconstruction, ensuring stable gradient flow and process reversibility. Furthermore, we introduce a novel fusion module called the Fusion-Aware Low-Rank Adaptation module, which dynamically calibrates spectral and spatial features while keeping the model lightweight. Extensive experiments on multiple benchmark datasets demonstrate that PIF-Net achieves significantly better image restoration performance than current state-of-the-art methods while maintaining model efficiency.

Spectral super-resolution, which reconstructs hyperspectral images (HSI) from a single RGB image, has garnered increasing attention. Due to the limitations of CNN structures in spectral modeling and the high computational cost of Transformer structures, existing deep learning (DL)-based methods struggle to balance spectral reconstruction quality and computational efficiency. Recently, Mamba methods base on state-space models (SSM) show great potential in modeling long-range dependencies with linear complexity. Therefore, we introduce the Mamba model into spectral super-resolution (SSR) task. Specifically, we propose a three-stage SSR network base on Mamba, called SSRMamba. We design SpaMamba, SSMamba, and SpeMamba modules for shallow spatial information extraction, mixed information encoding, and spectral information reconstruction, respectively. Extensive experimental results demonstrate that SSRMamba not only surpasses existing methods in terms of quantification and quality, achieving state-of-the-art (SOTA) performance, but also significantly reduces model size and computational cost. The source code of SSRMamba is available at: https://github.com/Baisonm-Li/SSRMamba.

Single hyperspectral image super-resolution (SHSR) aims to restore high-resolution images from low-resolution hyperspectral images. Recently, the Visual Mamba model has achieved an impressive balance between performance and computational efficiency. However, due to its 1D scanning paradigm, the model may suffer from potential artifacts during image generation. To address this issue, we propose HSRMamba. While maintaining the computational efficiency of Visual Mamba, we introduce a strip-based scanning scheme to effectively reduce artifacts from global unidirectional scanning. Additionally, HSRMamba uses wavelet decomposition to alleviate modal conflicts between high-frequency spatial features and low-frequency spectral features, further improving super-resolution performance. Extensive experiments show that HSRMamba not only excels in reducing computational load and model size but also outperforms existing methods, achieving state-of-the-art results. The open-source code is available at: https://github.com/oldsweet/HSRMamba.

Hyperspectral images (HSIs) have great potential in various visual tasks due to their rich spectral information. However, obtaining high-resolution hyperspectral images remains challenging due to limitations of physical imaging. Inspired by Kolmogorov-Arnold Networks (KANs), we propose an efficient HSI super-resolution (HSI-SR) model to fuse a low-resolution HSI (LR-HSI) and a high-resolution multispectral image (HR-MSI), yielding a high-resolution HSI (HR-HSI). To achieve the effective integration of spatial information from HR-MSI, we design a fusion module based on KANs, called KAN-Fusion. Further inspired by the channel attention mechanism, we design a spectral channel attention module called KAN Channel Attention Block (KAN-CAB) for post-fusion feature extraction. As a channel attention module integrated with KANs, KAN-CAB not only enhances the fine-grained adjustment ability of deep networks, enabling networks to accurately simulate details of spectral sequences and spatial textures, but also effectively avoid Curse of Dimensionality. Extensive experiments show that, compared to current state-of-the-art HSI-SR methods, proposed HSR-KAN achieves the best performance in terms of both qualitative and quantitative assessments. Our code is available at: https://github.com/Baisonm-Li/HSR-KAN.

Hyperspectral image super-resolution (SR) aims to reconstruct high-resolution images from low-resolution hyperspectral observations. Given the significant differences in feature dependencies across various frequency domains in SR tasks, some methods perform reconstruction in the frequency domain. However, directly applying Fourier or wavelet transforms to decompose features into fixed frequency components lacks adaptability, which limits the ability of the network to dynamically focus on high-value features. To address this limitation, this paper introduces a Dynamic Spectral Frequency Selection Model, which decomposes features into disentangled frequency sub-bands. Specifically, wavelet transform and wavelet decomposition are combined to achieve efficient frequency separation and selection, allowing the network to adaptively capture the most informative frequency components for reconstruction. The proposed Efficient Frequency Selective Network (EFSNet) achieves state-of-the-art performance in hyperspectral image SR, consistently surpassing existing methods on three publicly available datasets.

Resource management for Deep Learning (DL) clusters is essential for system efficiency and model training quality. Existing schedulers provided by DL frameworks are mostly adaptations from traditional HPC clusters and usually work on jobs’ makespan, assuming that DL training jobs finish completely. Unfortunately, it is reported that a fair amount of training jobs are exploratory jobs and often finish unsuccessfully (over 30%) in production clusters. This is due to the distinct characteristic of Deep Neural Network (DNN) training that it is an exploratory process of frequent user interventions, such as adjusting model structures, tuning hyperparameters, and exploring feature validity. Existing DL cluster schedulers using offline algorithms are not suitable for exploratory jobs when unexpected early terminations can cause noticeable resource waste. Moreover, DL training jobs are iterative and usually yield diminishing returns as they progress. Equally allocating resource among training iterations is not efficient, especially when dealing with exploratory jobs where it can worsen the degradation of system efficiency. The fundamental goal of a DL training job is to gain model quality improvement, usually indicated by the loss reduction (job profit) of a DNN model. This paper introduces a novel scheduling problem for exploratory jobs that seeks to maximize the overall training profit of a DL cluster. We propose ExplSched, an online scheduling solution based on the primal-dual framework, resulting in a competitive ratio of 2α that belongs to O(ln n). It uses a resource price function that emphasizes the importance of job profit to resource consumption ratio to make quick resource allocation decisions. Experimental results show that ExplSched achieved an average system utility improvement of 87.28% compared with other related work.