Modern datacenters and high-performance computing systems increasingly rely on SmartNICs to reduce host overhead and improve the security and efficiency of network data movement. This work-in-progress poster outlines our research path for a RISC-V-centric SmartNIC with dedicated context and objectives. The extended RISC-V core acts as the central controlling unit for the key SmartNIC components designed in RTL, including a security-oriented (Physically Unclonable Function) PUF unit for authentication, a flexible packet parser of arbitrary protocols enhanced by RISC-V Vector (RVV) extension, and a Remote Direct Memory Access (RDMA) engine targeted at high-performance computing clusters. We apply an end-to-end simulator to model these components at the behavioral level without requiring full hardware setups. Finally, this project maps the resulting design to an FPGA-based SmartNIC architecture.

Processor design and verification require a synergistic approach that combines instruction-level functional simulations with precise hardware emulations.The trade-off between speed and accuracy in the instruction set simulation poses a significant challenge to the efficiency of processor verification.By tapping the potentials of Field Programmable Gate Arrays (FPGAs), we propose an FPGA-assisted System-on-Chip (SoC) platform that facilitates cross-verification by the embedded CPU and the synthesized hardware in the programmable fabrics.This method accelerates the verification of the RISC-V Instruction Set Architecture (ISA) processor at a speed of 5 million instructions per second (MIPS), which is 150x faster than the vendor-specific tool (Xilinx XSim) and a 35x boost to the stateof-the-art open-source verification setup (Verilator).With less than 7% hardware occupation on Zynq 7000 FPGA, the proposed framework enables flexible verification with high time and cost efficiency for exploring RISC-V instruction set architectures.

With the rapid development of artificial intelligence technologies, the demand for training data has surged, exacerbating risks of data leakage. Despite increasing incidents and costs associated with such leaks, data leakage prevention (DLP) technologies lag behind evolving evasion techniques that bypass existing sensitive information recognition (SIR) models. Current datasets lack comprehensive coverage of these adversarial transformations, limiting the evaluation of robust SIR systems. To address this gap, we introduce DataSIR, a benchmark dataset specifically designed to evaluate SIR models on sensitive data subjected to diverse format transformations. We curate 26 sensitive data categories based on multiple international regulations, and collect 131,890 original samples correspondingly. Through empirical analysis of real-world evasion tactics, we implement 21 format transformation methods, which are applied to the original samples, expanding the dataset to 1,647,501 samples to simulate adversarial scenarios. We evaluated DataSIR using four traditional NLP models and four large language models (LLMs). For LLMs, we design structured prompts with varying degrees of contextual hints to assess the impact of prior knowledge on recognition accuracy. These evaluations demonstrate that our dataset effectively differentiates the performance of various SIR algorithms. Combined with its rich category and format diversity, the dataset can serve as a benchmark for evaluating related models and help develop future more advanced SIR models. Our dataset and experimental code are publicly available at https://www.kaggle.com/datasets/fanmo1/datasir and https://github.com/Fan-Mo-ZJU/DataSIR.

Quantum computing promises an effective way to solve targeted problems that are classically intractable. Among them, quantum computers built with superconducting qubits are considered one of the most advanced technologies, but they suffer from short coherence times. This can get exaggerated when they are controlled directly by general-purpose host machines, which in turn leads to the loss of quantum information. To mitigate this, we need quantum control processors (QCPs) positioned between quantum processing units (QPUs) and host machines to reduce latencies. However, existing QCPs are built on top of designs with no or inefficient scalability, requiring a large number of instructions when scaling to more qubits. In addition, interactions between current QCPs and host machines require frequent data transmissions and offline computations to obtain final results from hundreds of repeated executions, which limits the performance of quantum computers.In this paper, we propose a QCP — called HiSEP-Q — featuring a novel quantum instruction set architecture (QISA) and its microarchitecture implementation. For efficient control, we utilize mixed-type addressing modes and mixed-length instructions in HiSEP-Q, which provides an efficient way to concurrently address more than 100 qubits. Further, for efficient read-out and analysis, we develop a novel onboard accumulation and sorting unit, which eliminates the data transmission of raw data between the QCPs and host machines and enables real-time result processing. Compared to the state-of-the-art, our proposed QISA achieves at least 62% and 28% improvements in encoding efficiency with real and synthetic quantum circuits, respectively. We also validate the microarchitecture on a field-programmable gate array (FPGA), which exhibits low power and resource consumption, even as the number of qubits scales to 100. Both hardware and ISA evaluations demonstrate that HiSEP-Q features high scalability and efficiency toward the number of controlled qubits.

Aerial scene classification remains challenging as: 1) the size of key objects in determining the scene scheme varies greatly; 2) many objects irrelevant to the scene scheme are often flooded in the image. Hence, how to effectively perceive the region of interests (RoIs) from a variety of sizes and build more discriminative representation from such complicated object distribution is vital to understand an aerial scene. In this paper, we propose a novelall grains, one scheme(AGOS) framework to tackle these challenges.To the best of our knowledge, it is the first work to extend the classic multiple instance learning into multi-grain formulation. Specially, it consists of a multi-grain perception module (MGP), a multi-branch multi-instance representation module (MBMIR) and a self-aligned semantic fusion (SSF) module. Firstly, our MGP preserves the differential dilated convolutional features from the backbone, which magnifies the discriminative information from multi-grains. Then, our MBMIR highlights the key instances in the multi-grain representation under the MIL formulation. Finally, our SSF allows our framework to learn the same scene scheme from multi-grain instance representations and fuses them, so that the entire framework is optimized as a whole. Notably, our AGOS is flexible and can be easily adapted to existing CNNs in a plug-and-play manner. Extensive experiments on UCM, AID and NWPU benchmarks demonstrate that our AGOS achieves a comparable performance against the state-of-the-art methods.

The report introduces the specific content of the design and implementation of code-related practical courses by university teachers and industry professionals with profound software design and development experience. Based on scientific and engineering practice methodology and educational principles, the report introduces curriculum design principles, including the design principles of software engineering curriculum and the requirements of engineering accreditation. The report describes the course's timetable, course objectives, content, learning materials, learning outcomes, teacher activity and student activity, and assessment/evaluation. The report explains in detail how to help students develop the software engineering mindset based on industry best practice and real-world cases. The report summarizes the actual results of the course and proposes the next step to polish the course.

Aerial scene recognition is challenging due to the complicated object distribution and spatial arrangement in a large-scale aerial image. Recent studies attempt to explore the local semantic representation capability of deep learning models, but how to exactly perceive the key local regions remains to be handled. In this paper, we present a local semantic enhanced ConvNet (LSE-Net) for aerial scene recognition, which mimics the human visual perception of key local regions in aerial scenes, in the hope of building a discriminative local semantic representation. Our LSE-Net consists of a context enhanced convolutional feature extractor, a local semantic perception module and a classification layer. Firstly, we design a multi-scale dilated convolution operators to fuse multi-level and multi-scale convolutional features in a trainable manner in order to fully receive the local feature responses in an aerial scene. Then, these features are fed into our two-branch local semantic perception module. In this module, we design a context-aware class peak response (CACPR) measurement to precisely depict the visual impulse of key local regions and the corresponding context information. Also, a spatial attention weight matrix is extracted to describe the importance of each key local region for the aerial scene. Finally, the refined class confidence maps are fed into the classification layer. Exhaustive experiments on three aerial scene classification benchmarks indicate that our LSE-Net achieves the state-of-the-art performance, which validates the effectiveness of our local semantic perception module and CACPR measurement.

Deep learning methods have boosted the performance of a series of visual tasks. However, the aerial image scene classification remains challenging. The object distribution and spatial arrangement in aerial scenes are often more complicated than in natural image scenes. Possible solutions include highlighting local semantics relevant to the scene label and preserving more discriminative features. To tackle this challenge, in this letter, we propose an attention pooling-based dense connected convolutional network (APDC-Net) for aerial scene classification. First, it uses a simplified dense connection structure as the backbone to preserve features from different levels. Then, we propose a trainable pooling to down-sample the feature maps and to enhance the local semantic representation capability. Finally, we introduce a multi-level supervision strategy, so that features from different levels are all allowed to supervise the training process directly. Exhaustive experiments on three aerial scene classification benchmarks demonstrate that our proposed APDC-Net outperforms other state-of-the-art methods with much fewer parameters and validate the effectiveness of our attention-based pooling and multi-level supervision strategy.

The accuracy and efficiency of scene classification have immensely improved with the extensive application of deep convolutional neural networks (CNNs). However, standard CNNs classify images mostly based on the global features from the last fully connected layer, which may cause the negligence of discriminative local information and the sensitivity to various spatial transformations. In this article, we consider the problem of scene classification from the perspective of multiple instance learning (MIL) and propose an end-to-end multiple instance CNN (MI-CNN) for learning more robust scene representations. In MI-CNN, a scene is represented as a bag of local patches (instances). An instance-level classifier is trained to obtain the label of each patch in an MIL fashion, which makes the classifier more sensitive to the discriminative local patches. The patch labels are then aggregated into an image label by an MIL pooling layer, which is invariant to the order of local patches and helps construct more robust representations. We present extensive experiments on UC Merced Land use (UCM), Aerial Image data set (AID), and NWPU-RESISC (NWPU) data sets. Experimental results show that the proposed method achieves 1.17%, 1.70%, and 3.61% accuracy improvements with 90% parameter reduction compared with the standard CNNs.

In contrast with nature scenes, aerial scenes are often composed of many objects crowdedly distributed on the surface in bird's view, the description of which usually demands more discriminative features as well as local semantics. However, when applied to scene classification, most of the existing convolution neural networks (ConvNets) tend to depict global semantics of images, and the loss of low- and mid-level features can hardly be avoided, especially when the model goes deeper. To tackle these challenges, in this paper, we propose a multiple-instance densely-connected ConvNet (MIDC-Net) for aerial scene classification. It regards aerial scene classification as a multiple-instance learning problem so that local semantics can be further investigated. Our classification model consists of an instance-level classifier, a multiple instance pooling and followed by a bag-level classification layer. In the instance-level classifier, we propose a simplified dense connection structure to effectively preserve features from different levels. The extracted convolution features are further converted into instance feature vectors. Then, we propose a trainable attention-based multiple instance pooling. It highlights the local semantics relevant to the scene label and outputs the bag-level probability directly. Finally, with our bag-level classification layer, this multiple instance learning framework is under the direct supervision of bag labels. Experiments on three widely-utilized aerial scene benchmarks demonstrate that our proposed method outperforms many state-of-the-art methods by a large margin with much fewer parameters.

With the development of deep learning, many state-of-the-art natural image scene classification methods have demonstrated impressive performance. While the current convolution neural network tends to extract global features and global semantic information in a scene, the geo-spatial objects can be located at anywhere in an aerial image scene and their spatial arrangement tends to be more complicated. One possible solution is to preserve more local semantic information and enhance feature propagation. In this paper, an end to end multiple instance dense connected convolution neural network (MIDCCNN) is proposed for aerial image scene classification. First, a 23 layer dense connected convolution neural network (DCCNN) is built and served as a backbone to extract convolution features. It is capable of preserving middle and low level convolution features. Then, an attention based multiple instance pooling is proposed to highlight the local semantics in an aerial image scene. Finally, we minimize the loss between the bag-level predictions and the ground truth labels so that the whole framework can be trained directly. Experiments on three aerial image datasets demonstrate that our proposed methods can outperform current baselines by a large margin.

Spatial clustering can be used to discover hotspots in trajectory data. A trajectory clustering approach based on decision graph and data field is proposed as an effective method to select parameters for clustering, to determine the number of clusters, and to identify cluster centers. Synthetic data and real-world taxi trajectory data are utilized to demonstrate the effectiveness of the proposed approach. Results show that the proposed method can automatically determine the parameters for clustering as well as perform efficiently in trajectory clustering. Hotspots are identified and visualized during different times of a single day and at the same times on different days. The dynamic patterns of hotspots can be used to identify crowded areas and events, which are crucial for urban transportation planning and management.

Built-up areas are typical man-made structure in urban environment, and timely and accurately acquiring built-up area layers can provide necessary geo-spatial information for planners and policymakers. In this paper, a data field-based method is proposed for the automated detection of built-up areas from high-resolution satellite images. This method views the local corner features of buildings as mass points and models their spatial interaction and distribution by potential function. Due to the salient potential differences between built-up areas and non-built-up areas, the built-up areas are extracted by threshold based segmentation. Two further post-processing techniques, that is, noise removal and hole filling, significantly improve the detection results. The experimental results indicate that the proposed method shows very good performance, and it, with simple parameter settings and without needing sample information, not only can achieve high extraction accuracy but also can effectively keep the shape and topological structure of built-up areas by further post-processing techniques.

In the classification of high-resolution remote sensing images, spatial correlations between pixel values are important spatial information. Traditional methods of measuring spatial correlation are inadequate for the extraction of spatial information about the shape and structure of object classes. In this letter, we propose a novel method using directional spatial correlation (DSC) to model and extract spatial information in neighborhoods of pixels. Two sets of descriptors DSC_I and DSC_C are defined to describe spatial structural features. The effectiveness of the proposed method was tested by image classification on two data sets. Results show that the DSC-based approach can drastically improve the classification, and it is found by comparison that it has better performance than some existing methods.

The time series remote sensing data and meteorological satellite data offer new opportunities for understanding the earth system. Spatio-temporal data clustering becomes a kind of idea tool to explore huge data space of spatio-temporal data. Because there are many uncertainties in the huge spatio-temporal data, including fuzziness and randomness, the spatio-temporal clustering methods with uncertainties are needed. Based on type-2 fuzzy sets and cloud models, the paper analyzes the uncertainty of the membership of FCM (fuzzy C-means), and proposes CFFCM (cloud fuzzifier fuzzy C-means) method. Take the time series SST (sea surface temperature) data as examples, the paper applies CFFCM to carry out spatio-temporal clustering analysis, and discovers some interesting patterns.