Despite being empowered with alignment mechanisms, large language models (LLMs) are increasingly vulnerable to emerging jailbreak attacks that can compromise their alignment mechanisms. This vulnerability poses significant risks to real-world applications. Existing work faces challenges in both training efficiency and generalization capabilities (i.e., Reinforcement Learning from Human Feedback and Red-Teaming). Developing effective strategies to enable LLMs to resist continuously evolving jailbreak attempts represents a significant challenge. To address this challenge, we propose a novel defensive paradigm called GuidelineLLM, which assists LLMs in recognizing queries that may have harmful content. Before LLMs respond to a query, GuidelineLLM first identifies potential risks associated with the query, summarizes these risks into guideline suggestions, and then feeds these guidelines to the responding LLMs. Importantly, our approach eliminates the necessity for additional safety fine-tuning of the LLMs themselves; only the GuidelineLLM requires fine-tuning. This characteristic enhances the general applicability of GuidelineLLM across various LLMs. Experimental results demonstrate that GuidelineLLM can significantly reduce the attack success rate (ASR) against LLM (an average reduction of 34.17% ASR) while maintaining the usefulness of LLM in handling benign queries.

As a longstanding scientific challenge, accurate and timely ocean forecasting has always been a sought-after goal for ocean scientists. However, traditional theory-driven numerical ocean prediction (NOP) suffers from various challenges, such as the indistinct representation of physical processes, inadequate application of observation assimilation, and inaccurate parameterization of models, which lead to difficulties in obtaining effective knowledge from massive observations, and enormous computational challenges. With the successful evolution of data-driven deep learning in various domains, it has been demonstrated to mine patterns and deep insights from the ever-increasing stream of oceanographic spatiotemporal data, which provides novel possibilities for revolution in ocean forecasting. Deep-learning-based ocean forecasting (DLOF) is anticipated to be a powerful complement to NOP. Nowadays, researchers attempt to introduce deep learning into ocean forecasting and have achieved significant progress that provides novel motivations for ocean science. This article provides a comprehensive review of the state-of-the-art DLOF research regarding model architectures, spatiotemporal multiscales, and interpretability while specifically demonstrating the feasibility of developing hybrid architectures that incorporate theory-driven and data-driven models. Moreover, we comprehensively evaluate DLOF from datasets, benchmarks, and cloud computing. Finally, the limitations of current research and future trends of DLOF are also discussed and prospected.

Precipitation nowcasting pertains to the localized forecasting of rainfall over a brief time horizon, characterized by precise estimates of both coverage and intensity. This capability holds particular significance in various societal applications, including agriculture, aviation safety, and transportation. However, since traditional methods mainly by extrapolating radar echo in 2-D space, cannot accurately and sufficiently represent the spatiotemporal state of clouds in the vertical direction, the accuracy of precipitation nowcasting using weather radar has reached a bottleneck. A new deep learning precipitation nowcasting model called 3dCloudNet is designed and evaluated in this study. The 3dCloudNet incorporates historical 3-D radar echo sequences obtained from weather radar data to improve the accuracy and reliability of precipitation nowcasting. By capturing both horizontal and vertical motion patterns of clouds at various altitude levels, this model demonstrates an enhanced capability in detecting and distinguishing regions prone to severe convective weather events. The experimental results show that the model better captures the cloud’s motion patterns and trends, and therefore has a noteworthy ability to detect and distinguish areas that may lead to severe convective weather. This study provides a step toward further improving the accuracy of precipitation nowcasting.

High-resolution land cover mapping (LCM) is an important application for studying and understanding the change of the earth surface. While deep learning (DL) methods demonstrate great potential in analyzing satellite images, they largely depend on massive high-quality labels. This paper proposes SW-LCM, a Scalable and Weakly-supervised two-stage Land Cover Mapping method on a new Sunway Supercomputer. Our method consists of a k-means clustering module as a first stage, and an iterative deep learning module as a second stage. With the k-means module providing a good enough starting point (taking inaccurate results as noisy labels), the deep learning module improves the classification results in an iterative way, without any labelling efforts required for processing large scenarios. To achieve efficiency for country-level land cover mapping, we design a customized data partition scheme and an on-the-fly assembly for k-means. Through careful parallelization and optimization, our k-means module scales to 98,304 computing nodes (over 38 million cores), and provides a sustained performance of 437.56 PFLOPS, in a real LCM task of the entire region of China; the iterative updating part scales to 24,576 nodes, with a performance of 11 PFLOPS. We produce a 10-m resolution land cover map of China, with an accuracy of 83.5% (10-class) or 73.2% (25-class), 7% to 8% higher than best existing products, paving ways for finer land surveys to support sustainability-related applications.

Due to the ability of feature extraction, deep learning (DL)-based methods have been recently applied to channel state information (CSI) compression feedback in massive multiple-input multiple-output (MIMO) systems. Existing DL-based CSI compression methods are usually effective in extracting a certain type of features in the CSI. However, the CSI usually contains two types of propagation features, i.g., non-line-of-sight (NLOS) propagation-path feature and dominant propagation-path feature, especially in channel environments with rich scatterers. To fully extract the both propagation features and learn a dual-feature representation for CSI, this paper proposes a dual-feature-fusion neural network (NN), referred to as DuffinNet. The proposed DuffinNet adopts a parallel structure with a convolutional neural network (CNN) and an attention-empowered neural network (ANN) to respectively extract different features in the CSI, and then explores their interplay by a fusion NN. Built upon this proposed DuffinNet, a new encoder-decoder framework is developed, referred to as Duffin-CsiNet, for improving the end-to-end performance of CSI compression and reconstruction. To facilitate the application of Duffin-CsiNet in practice, this paper also presents a two-stage approach for codeword quantization of the CSI feedback. Besides, a transfer learning-based strategy is introduced to improve the generalization of Duffin-CsiNet, which enables the network to be applied to new propagation environments. Simulation results illustrate that the proposed Duffin-CsiNet noticeably outperforms the existing DL-based methods in terms of reconstruction performance, encoder complexity, and network convergence, validating the effectiveness of the proposed dual-feature fusion design.

In order to fully exploit the advantages of massive multiple-input multiple-output (mMIMO), it is critical for the transmitter to accurately acquire the channel state information (CSI). Deep learning (DL)-based methods have been proposed for CSI compression and feedback to the transmitter. Although most existing DL-based methods consider the CSI matrix as an image, structural features of the CSI image are rarely exploited in neural network design. As such, we propose a model of self-information that dynamically measures the amount of information contained in each patch of a CSI image from the perspective of structural features. Then, by applying the self-information model, we propose a model-and-data-driven network for CSI compression and feedback, namely IdasNet. The IdasNet includes the design of a module of self-information deletion and selection (IDAS), an encoder of informative feature compression (IFC), and a decoder of informative feature recovery (IFR). In particular, the model-driven module of IDAS pre-compresses the CSI image by removing informative redundancy in terms of the self-information. The encoder of IFC then conducts feature compression to the pre-compressed CSI image and generates a feature codeword which contains two components, i.e., codeword values and position indices of the codeword values. Subsequently, the IFR decoder decouples the codeword values as well as position indices to recover the CSI image. Experimental results verify that the proposed IdasNet noticeably outperforms existing DL-based networks under various compression ratios while it has the number of network parameters reduced by orders-of-magnitude compared with various existing methods.

The increasing complexity of data analysis tasks makes it dependent on human expertise and challenging for non-experts. One of the major challenges faced in data analysis is the selection of the proper algorithm for given tasks and data sets. Motivated by this, we develop Assassin, aiming at helping users without enough expertise to automatically select optimal algorithms for classification tasks. By embedding meta-learning techniques and reinforced policy, our system can automatically extract experiences from previous tasks and train a meta-classifier to implement algorithm recommendations. Then we apply genetic search to explore hyperparameter configuration for the selected algorithm. We demonstrate Assassin with classification tasks from OpenML. The system chooses an appropriate algorithm and optimal hyperparameter configuration for them to achieve a high-level performance target. The Assassin has a user-friendly interface that allows users to customize the parameters during the search process.