Search for: All records

Abstract Prescribed burning is an essential forest management tool that requires strategic planning to effectively address its multidimensional impacts, particularly given the influence of global climate change on fire behavior. Despite the inherent complexity in planning prescribed burns, limited efforts have been made to comprehensively identify the critical elements necessary for formulating effective models. In this work, we present a systematic review of the literature on optimization and decision models for prescribed burning, analyzing 471 academic papers published in the last 25 years. Our study identifies four main types of models: spatial‐allocation, spatial‐extent, temporal‐only, and spatial–temporal. We observe a growing number of studies on modeling prescribed burning, primarily due to the expansion in spatial‐allocation and spatial–temporal models. There is also an increase in complexity as the models consider more elements affecting prescribed burning effectiveness. We identify the essential components for optimization models, including stakeholders, decision variables, objectives, and influential factors, to enhance model practicality. The review also examines solution techniques, such as integer programming in spatial allocation, stochastic dynamic programming in probabilistic models, and multiobjective programming in balancing trade‐offs. These techniques' strengths and limitations are discussed to help researchers adapt methods to specific challenges in prescribed burning optimization. In addition, we investigate general assumptions in the models and challenges in relaxation to enhance practicality. Lastly, we propose future research to develop more comprehensive models incorporating dynamic fire behaviors, stakeholder preferences, and long‐term impacts. Enhancing these models' accuracy and applicability will enable decision‐makers to better manage wildfire treatment outcomes. 

With the growing development and deployment of large language models (LLMs) in both industrial and academic fields, their security and safety concerns have become increasingly critical. However, recent studies indicate that LLMs face numerous vulnerabilities, including data poisoning, prompt injections, and unauthorized data exposure, which conventional methods have struggled to address fully. In parallel, blockchain technology, known for its data immutability and decentralized structure, offers a promising foundation for safeguarding LLMs. In this survey, we aim to comprehensively assess how to leverage blockchain technology to enhance LLMs' security and safety. Besides, we propose a new taxonomy of blockchain for large language models (BC4LLMs) to systematically categorize related works in this emerging field. Our analysis includes novel frameworks and definitions to delineate security and safety in the context of BC4LLMs, highlighting potential research directions and challenges at this intersection.Through this study, we aim to stimulate targeted advancements in blockchain-integrated LLM security. 

In recent years, plentiful evidence illustrates that Graph Con- volutional Networks (GCNs) achieve extraordinary accom- plishments on the node classification task. However, GCNs may be vulnerable to adversarial attacks on label-scarce dy- namic graphs. Many existing works aim to strengthen the ro- bustness of GCNs; for instance, adversarial training is used to shield GCNs against malicious perturbations. However, these works fail on dynamic graphs for which label scarcity is a pressing issue. To overcome label scarcity, self-training attempts to iteratively assign pseudo-labels to highly confi- dent unlabeled nodes but such attempts may suffer serious degradation under dynamic graph perturbations. In this paper, we generalize noisy supervision as a kind of self-supervised learning method and then propose a novel Bayesian self- supervision model, namely GraphSS, to address the issue. Extensive experiments demonstrate that GraphSS can not only affirmatively alert the perturbations on dynamic graphs but also effectively recover the prediction of a node classifier when the graph is under such perturbations. These two advan- tages prove to be generalized over three classic GCNs across five public graph datasets. 

Supervised learning, while deployed in real-life scenarios, often encounters instances of unknown classes. Conventional algorithms for training a supervised learning model do not provide an option to detect such instances, so they miss-classify such instances with 100% probability. Open Set Recognition (OSR) and Non-Exhaustive Learning (NEL) are potential solutions to overcome this problem. Most existing methods of OSR first classify members of existing classes and then identify instances of new classes. However, many of the existing methods of OSR only makes a binary decision, i.e., they only identify the existence of the unknown class. Hence, such methods cannot distinguish test instances belonging to incremental unseen classes. On the other hand, the majority of NEL methods often make a parametric assumption over the data distribution, which either fail to return good results, due to the reason that real-life complex datasets may not follow a well-known data distribution. In this paper, we propose a new online non-exhaustive learning model, namely, Non-Exhaustive Gaussian Mixture Generative Adversarial Networks (NE-GM-GAN) to address these issues. Our proposed model synthesizes Gaussian mixture based latent representation over a deep generative model, such as GAN, for incremental detection of instances of emerging classes in the test data. Extensive experimental results on several benchmark datasets show that NE-GM-GAN significantly outperforms the state-of-the-art methods in detecting instances of novel classes in streaming data.