Reinforcement learning has emerged as a paradigm for post-training large language models, boosting their reasoning capabilities. Such approaches compute an advantage value for each sample, reflecting better or worse performance than expected, thereby yielding both positive and negative signals for training. However, existing approaches often mix them indiscriminately, especially in the early stages, leading to ambiguous guidance and limited gains. To address this issue, we propose CAPO (Curriculum Advantage Policy Optimization), an adaptive curriculum mechanism based on advantage signals. The proposed mechanism bootstraps imitation learning with positive-only advantage samples to establish robust foundations, and subsequently introduces negative signals to cultivate discriminative capabilities, thereby improving generalization across complex scenarios. Compatible with diverse optimization methods including GRPO, PPO, RLOO, and Reinforce++, our method consistently achieves stable and significant improvements in mathematical reasoning tasks, and further generalizes effectively to multimodal Graphical User Interface (GUI) reasoning scenarios, establishing itself as a versatile and robust optimization paradigm.

The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships. Existing benchmarks usually adopt notably smaller image sizes than real-world RS scenarios, suffer from limited annotation quality, and consider insufficient dimensions of evaluation. To address these issues, we present XLRS-Bench: a comprehensive benchmark for evaluating the perception and reasoning capabilities of MLLMs in ultra-high-resolution RS scenarios. XLRS-Bench boasts the largest average image size (8500×8500) observed thus far, with all evaluation samples meticulously annotated manually, assisted by a novel semi-automatic captioner on ultra-high-resolution RS images. On top of the XLRS-Bench, 16 sub-tasks are defined to evaluate MLLMs’ 10 kinds of perceptual capabilities and 6 kinds of reasoning capabilities, with a primary emphasis on advanced cognitive processes that facilitate real-world decision-making and the capture of spatiotemporal changes. The results of both general and RS-focused MLLMs on XLRS-Bench indicate that further efforts are needed for real-world RS applications. We have open-sourced XLRS-Bench to support further research in developing more powerful MLLMs for remote sensing.

Ultra-high-resolution (UHR) remote sensing (RS) imagery offers valuable data for Earth observation but pose challenges for existing multimodal foundation models due to two key bottlenecks: (1) limited availability of UHR training data, and (2) token explosion caused by the large image size. To address data scarcity, we introduce **SuperRS-VQA** (avg. 8,376$\times$8,376) and **HighRS-VQA** (avg. 2,000$\times$1,912), the highest-resolution vision-language datasets in RS to date, covering 22 real-world dialogue tasks. To mitigate token explosion, our pilot studies reveal significant redundancy in RS images: crucial information is concentrated in a small subset of object-centric tokens, while pruning background tokens (e.g., ocean or forest) can even improve performance. Motivated by these findings, we propose two strategies: *Background Token Pruning* and *Anchored Token Selection*, to reduce the memory footprint while preserving key semantics. Integrating these techniques, we introduce **GeoLLaVA-8K**, the first RS-focused multimodal large language model capable of handling inputs up to 8K$\times$8K resolution, built on the LLaVA framework. Trained on SuperRS-VQA and HighRS-VQA, GeoLLaVA-8K sets a new state-of-the-art on the XLRS-Bench. Datasets and code were released at https://github.com/MiliLab/GeoLLaVA-8K.

Recent advances in self-supervised learning for Vision Transformers (ViTs) have fueled breakthroughs in remote sensing (RS) foundation models. However, the quadratic complexity of self-attention poses a significant barrier to scalability, particularly for large models and high-resolution images. While the linear-complexity Mamba architecture offers a promising alternative, existing RS applications of Mamba remain limited to supervised tasks on small, domain-specific datasets. To address these challenges, we propose RoMA, a framework that enables scalable self-supervised pretraining of Mamba-based RS foundation models using large-scale, diverse, unlabeled data. RoMA enhances scalability for high-resolution images through a tailored auto-regressive learning strategy, incorporating two key innovations: 1) a rotation-aware pretraining mechanism combining adaptive cropping with angular embeddings to handle sparsely distributed objects with arbitrary orientations, and 2) multi-scale token prediction objectives that address the extreme variations in object scales inherent to RS imagery. Systematic empirical studies validate that Mamba adheres to RS data and parameter scaling laws, with performance scaling reliably as model and data size increase. Furthermore, experiments across scene classification, object detection, and semantic segmentation tasks demonstrate that RoMA-pretrained Mamba models consistently outperform ViT-based counterparts in both accuracy and computational efficiency. The source code and pretrained models have be released at https://github.com/MiliLab/RoMA.

Shipboard integrated power systems (SIPSs) are usually equipped with multiple pulsed power loads (PPLs). In complex scenarios with strict operational requirements, the performance of the SIPS is greatly affected by the output energy of PPLs, especially in achieving high suppressive capability in a short period. To maximize the short-time output efficiency of PPLs in emergency situations, the aperiodic scheduling method is studied, aiming at optimizing the coordination of multiple PPLs and energy storage. The proposed aperiodic scheduling model aims to maximize the total utility of PPLs in the given period. The model fully considers constraints related to energy storage, system power balance, aperiodic charging and discharging process, time sequence, and charging power. To solve this complex non-linear model, a bi-layer dynamic programming algorithm is proposed. Simulation tests of SIPS performance are carried out to validate the advantage of the proposed method. The results show that the performance of single PPL and multi PPLs with aperiodic scheduling is improved by 42.3% and 25.7% respectively compared with periodic scheduling.

Abstract The height measurement algorithm of metre wave multiple input multiple output (MIMO) radar based on uniform linear array (ULA) has some problems, such as a sharp decline of angle measurement accuracy of ultra‐low altitude targets and large fluctuation of height measurement with the change of elevation. In order to solve the above difficult problems, starting from the perspective of reducing the influence of multipath effect by array structure, coprime array (CPA) is applied to the height measurement of monostatic metre wave MIMO radar. In this paper, the signal model of monostatic CPA metre wave MIMO radar is established under the condition of flat ground, a height measurement method using physical array is proposed based on the derivation that it is infeasible to use virtual array method for height measurement, it fills the gap of low elevation estimation method based on sparse array metre wave MIMO radar. The proposed method calculates the echo data according to the characteristics of the steering vector of CPA, then obtains the accurate low elevation by using the generalised multiple signal classification algorithm, and finally the target height is obtained by using the geometric relationship. Simulation experiments compare three typical CPAs with ULA using equal array elements, and verify the superiority of height measurement performance of monostatic metre wave MIMO radar based on CPA and the effectiveness of the proposed method.

Abstract The tensor‐based spatial smoothing is a good algorithm for the direction of arrival (DOA) estimation, but it cannot make good use of the physical aperture of the array, so the paper proposes a tensor‐based array virtual translation DOA estimation algorithm to solve the problem. Under the framework of tensor‐based DOA estimation algorithm, the factor matrix obtained by tensor decomposition is extended to a signal subspace with Vandermonde structure by using array virtual translation technology. In addition, the algorithm expands the available array aperture, breaks through the limitation of physical array aperture on multi‐target estimation ability, and further improves the estimation accuracy. Since the processing technique proposed in this paper has nothing to do with the construction of tensors, this technique is suitable for conventional DOA estimation algorithms based on tensors. Theoretical analysis and simulation experiments verify the effectiveness of the algorithm proposed in this paper.

In this paper, we propose a novel approach of video text keyframe detection, to achieve the goal of representing the video with textual keyframes and reducing the waste of resources in video review. Different from the works of video summarization which mainly focus on the variation of the scenes in videos, we pay attention to the variances of the text between sequential frames. For the above purpose, Text Siamese Network (TSN) is developed to automatically detect the keyframes which contain text in videos. Specifically, the TSN is composed of two branches, text similarity measurement and text identification. The first branch is utilized to evaluate the similarity between consecutive frames. Furthermore, an attention block is used to select the informative features to identify whether a frame contains text or not in the second branch. Additionally, a new dataset called VTKD2019 is proposed for video text keyframe detection. VTKD2019 contains 571 videos and is spit into three levels (easy, medium and hard) for evaluation. The experimental results on the VTKD2019 and ICDAR2015 demonstrate the effectiveness of our method.

Chinese keyword spotting is a challenging task as there is no visual blank for Chinese words. Different from English words which are split naturally by visual blanks, Chinese words are generally split only by semantic information. In this paper, we propose a new Chinese keyword spotter for natural images, which is inspired by Mask R-CNN. We propose to predict the keyword masks guided by text line detection. Firstly, proposals of text lines are generated by Faster R-CNN; Then, text line masks and keyword masks are predicted by segmentation in the proposals. In this way, the text lines and keywords are predicted in parallel. We create two Chinese keyword datasets based on RCTW-17 and ICPR MTWI2018 to verify the effectiveness of our method.

Accurate road detection and centerline extraction from very high resolution (VHR) remote sensing imagery are of central importance in a wide range of applications. Due to the complex backgrounds and occlusions of trees and cars, most road detection methods bring in the heterogeneous segments; besides for the centerline extraction task, most current approaches fail to extract a wonderful centerline network that appears smooth, complete, as well as single-pixel width. To address the above-mentioned complex issues, we propose a novel deep model, i.e., a cascaded end-to-end convolutional neural network (CasNet), to simultaneously cope with the road detection and centerline extraction tasks. Specifically, CasNet consists of two networks. One aims at the road detection task, whose strong representation ability is well able to tackle the complex backgrounds and occlusions of trees and cars. The other is cascaded to the former one, making full use of the feature maps produced formerly, to obtain the good centerline extraction. Finally, a thinning algorithm is proposed to obtain smooth, complete, and single-pixel width road centerline network. Extensive experiments demonstrate that CasNet outperforms the state-of-the-art methods greatly in learning quality and learning speed. That is, CasNet exceeds the comparing methods by a large margin in quantitative performance, and it is nearly 25 times faster than the comparing methods. Moreover, as another contribution, a large and challenging road centerline data set for the VHR remote sensing image will be publicly available for further studies.

Building extraction from remote sensing images is of great importance in urban planning. Yet it is a longstanding problem for many complicate factors such as various scales and complex backgrounds. This paper proposes a novel supervised building extraction method via deep deconvolution neural networks (DeconvNet). Our method consists of three steps. First, we preprocess the multi-source remote sensing images provided by the IEEE GRSS Data Fusion Contest. A high-quality Vancouver building dataset is created on pansharpened images whose ground-truth are obtained from the OpenStreetMap project. Then, we pretrain a deep deconvolution network on a public large-scale Massachusetts building dataset, which is further fine-tuned by two band combinations (RGB and NRG) of our dataset, respectively. Moreover, the output saliency maps of the fine-tuned models are fused to produce the final building extraction result. Extensive experiments on our Vancouver building dataset demonstrate the effectiveness and efficiency of the proposed method. To the best of our knowledge, it is the first work to use deconvolution networks for building extraction from remote sensing images.

Extracting ships from complex backgrounds is the bottleneck of ship detection in high-resolution optical satellite images. In this letter, we propose a nearly closed-form ship rotated bounding box space used for ship detection and design a method to generate a small number of highly potential candidates based on this space. We first analyze the possibility of accurately covering all ships by labeling rotated bounding boxes. Moreover, to reduce search space, we construct a nearly closed-form ship rotated bounding box space. Then, by scoring for each latent candidate in the space using a two-cascaded linear model followed by binary linear programming, we select a small number of highly potential candidates. Moreover, we also propose a fast version of our method. Experiments on our data set validate the effectiveness of our method and the efficiency of its fast version, which achieves a close detection rate in near real time.

Generating object proposals before object detection has become a common way. In this paper, we present a novel method to measure the objectness of bounding boxes using edges. The contours play an important role in object localization and detection. The number of edges that are close to the boundary of a box has strong relationship with the likelihood of the box covering an object. In our method, we adopt a two-step scheme to generate object proposals. In the first step, we count the number of contours close to the box, where we use the proposed “Tile Algorithm” to wipe off the inner edges of a box. In the second step we re-rank the object proposals with a linear SVM classifier across all aspect-ratios for calibration. Experiments on the VOC2007 dataset show that we achieve 96.47% object detection rate with 1000 proposals.