Software-defined networking and network function virtualization have brought unparalleled flexibility in defining and managing network architectures. With the widespread diffusion of cloud platforms, more resources ar...
详细信息
ISBN:
(纸本)9789819608072;9789819608089
Software-defined networking and network function virtualization have brought unparalleled flexibility in defining and managing network architectures. With the widespread diffusion of cloud platforms, more resources are available to execute virtual network functions concurrently, but the current approach to defining networks in the cloud development is held back by the lack of tools to manage the composition of more complex flows than simple sequential invocations. In this paper, we advocate for the usage of choreographic programming for defining the multiparty workflows of a network. When applied to the composition of virtual network functions, this approach yields multiple advantages: a single program expresses the behavior of all components, in a way that is easier to understand and check;a compiler can produce the executable code for each component, guaranteeing correctness properties of their interactions such as deadlock freedom;and the bottleneck of a central orchestrator is removed. We describe the proposed approach and show its feasibility via a case study where different functions cooperatively solve a security monitoring task.
Feature-based knowledge distillation has been recognized a remarkably effective way to transfer informative knowledge from a complicated teacher model to a simple student model. However, for most knowledge distillatio...
详细信息
ISBN:
(纸本)9789819785018;9789819785025
Feature-based knowledge distillation has been recognized a remarkably effective way to transfer informative knowledge from a complicated teacher model to a simple student model. However, for most knowledge distillation methods, the teacher model merely regards the feature knowledge as a supervisory information but neglects its guidance to the student model, leading to a large gap between the feature knowledge of the teacher and that of the student. To overcome this weakness, we propose a novel preliminary knowledge guided distillation that incorporates the layer-level features from the teacher as prior knowledge to guide the student to generate the guided features. The guided features can narrow the difference between the teacher knowledge and the student knowledge. Furthermore, to enhance the quality of teacher features, a Multi-Level Feature Fusion module is employed to integrate the rich context of the teacher features across different levels, which benefits to more comprehensive exploration of teacher features. We validate the superiority of our approach by performing experiments on three different tasks, i.e., Image Classification on CIFAR-100 and Tiny ImageNet datasets, Object Detection and Instance Segmentation on MSCOCO dataset, respectively, indicating more competitive performance than other typical approaches.
Text-Image relation inference (TIRI) aims to identify the potential semantic relationships between text and image. Although previous works have made some progress, there are still two issues that are worth exploring. ...
详细信息
ISBN:
(纸本)9789819755547;9789819755554
Text-Image relation inference (TIRI) aims to identify the potential semantic relationships between text and image. Although previous works have made some progress, there are still two issues that are worth exploring. First, existing studies primarily rely on the English text and the accompanying image to perform TIRI. This completely ignores other external information (e.g, another parallel the linguistic knowledge from Chinese by machine translation). As previous studies, bilingual knowledge can help us understand the semantics of the text more accurately, like polysemy problems. Second, existing studies normally employ a Transformer-based structure and implicitly encode different modalities. This completely neglects the potential dependencies among each unit in the uni-modality (e.g., the dependency syntax in the text). Therefore, we propose a bilingual multimodal graph convolutional network (BMGCN) to model both intra-modal and inter-modal dependence in a fine-grained manner. This approach can not only explicitly model the dependencies within each modality and each language, but also model the dependencies between different modalities and different languages simultaneously. Systematic experiments demonstrate that our BMGCN obviously outperforms the state-of-the-art on two datasets. Additionally, we provide several interesting analyses to further verify the effectiveness of our proposed approach.
Currently, large language models (LLMs) are the state of the art for pre-trained language models. LLMs have been applied to many tasks, including question and answering over Knowledge Graphs (KGs) and text-to-SPARQL, ...
详细信息
ISBN:
(纸本)9783031789519;9783031789526
Currently, large language models (LLMs) are the state of the art for pre-trained language models. LLMs have been applied to many tasks, including question and answering over Knowledge Graphs (KGs) and text-to-SPARQL, that is, the translation of Natural Language (NL) questions to SPARQL queries. This paper introduces Auto-KGQA, an autonomous domain-independent framework based on LLMs for text-to-SPARQL. The framework uses as context, fragments of the KG, which the LLM uses to translate the user's NL question to a SPARQL query on the KG. Finally, it generates a natural language response for the user, based upon the result of the execution of SPARQL query over the KG.
Attack Trees are a graphical model of security used to study threat scenarios. While visually appealing and supported by solid theories and effective tools, one of their main drawbacks remains the amount of effort req...
详细信息
ISBN:
(纸本)9783031737084;9783031737091
Attack Trees are a graphical model of security used to study threat scenarios. While visually appealing and supported by solid theories and effective tools, one of their main drawbacks remains the amount of effort required by security experts to design them from scratch. This work aims at remedying this by providing a method for the automatic generation of Attack Trees from attack logs. The main original feature of our approach w.r.t. existing ones is the use of Process Mining algorithms to synthesize Attack Trees, which allow users to customize the way a set of logs are summarized as an Attack Tree, for example by discarding statistically irrelevant events. Our approach is supported by a prototype that, apart from the derivation and translation of the model, provides the user with an Attack Tree in the RisQFLan format, a tool used for quantitative risk modeling and analysis with Attack Trees. We use literature case studies to illustrate and explore the capabilities of our approach.
Model checking trees generated by Higher-Order Recursion Schemes (HORS) of order k against Alternating Parity Tree-Automata (APT) is known to be a k-EXPTIME-complete problem (Ong'06). We exhibit a natural fragment...
详细信息
ISBN:
(纸本)9783031826993;9783031827006
Model checking trees generated by Higher-Order Recursion Schemes (HORS) of order k against Alternating Parity Tree-Automata (APT) is known to be a k-EXPTIME-complete problem (Ong'06). We exhibit a natural fragment of HORS, called tail-recursive HORS, and a restricted APT model, called bounded-alternation APT, such that the problem of model checking trees generated by order-k tail-recursive HORS against bounded-alternation APT is k-1-EXPSPACE-complete. The upper bound is achieved by converting the problem into an alternating reachability game, the lower one via reduction from a tiling problem.
Recently, the importance of generating accurate question and answer summaries has increased because the number of documents has increased, but it is difficult to summarize them from unstructured documents where the an...
详细信息
ISBN:
(纸本)9789819609130;9789819609147
Recently, the importance of generating accurate question and answer summaries has increased because the number of documents has increased, but it is difficult to summarize them from unstructured documents where the answer is not associated with the question. To address this problem, first we make a group of answer sentences, which corresponds to one question based on the heuristic rules, and we find the corresponding answer from the summaries and subtopics of the question by using the BM25-based similarity calculation. Second, we use the large language model (LLM) to generate a summary of the answers from the answers found. Experimental results showed that our methods significantly outperformed LLM-based answer generation that inputs whole answer sentences including irrelevant parts, which correspond to another question, to LLM and that our methods were practical compared to human-generated gold summaries.
Multiparty session typing (MPST) is a formal method to make concurrent programming simpler. The idea is to use type checking to automatically prove safety (protocol compliance) and liveness (communication deadlock fre...
详细信息
ISBN:
(纸本)9783031711763;9783031711770
Multiparty session typing (MPST) is a formal method to make concurrent programming simpler. The idea is to use type checking to automatically prove safety (protocol compliance) and liveness (communication deadlock freedom) of implementations relative to specifications. Discourje is an existing run-time verification library for communication protocols in Clojure, based on dynamic MPST. The original version of Discourje can detect only safety violations. In this paper, we present an extension of Discourje to detect also liveness violations.
Semantic image synthesis, involves the transformation of semantic layouts into realistic images, is aimed at comprehending and leveraging given semantic information. Despite recent impressive advancements, challenges ...
详细信息
ISBN:
(纸本)9789819785049;9789819785056
Semantic image synthesis, involves the transformation of semantic layouts into realistic images, is aimed at comprehending and leveraging given semantic information. Despite recent impressive advancements, challenges persist in terms of fidelity, semantic alignment, and training stability. To enhance the generation quality and semantic alignment in semantic image synthesis, we have reengineered the noise mapping and semantic space embedding, proposing a novel semantic image synthesis model, GAN-Diffusion Relay Model (GDRM), based on GAN and relay diffusion model. Extensive experiments on benchmark datasets validate the effectiveness of our proposed approach, achieving state-of-the-art performance in terms of fidelity (FID) and diversity (LPIPS).
The road crack detection remains a crucial task in the road maintenance and safety management. However, due to the diversity and complexity of cracks, achieving the fine-grained and accurate segmentation is still chal...
详细信息
ISBN:
(纸本)9789819785018;9789819785025
The road crack detection remains a crucial task in the road maintenance and safety management. However, due to the diversity and complexity of cracks, achieving the fine-grained and accurate segmentation is still challenging. To this end, this paper proposes a novel physically informed prior-guided crack segmentation method. Specifically, we employ the dynamic snake convolution to enhance the segmentation continuity and consistency. Moreover, a prior information is injected to supplement the morphology and structural features of road cracks, aiming to mitigate the miss detection of the binary-branched and webbed cracks. To ensure the continuity and completeness of cracks, a cross-correlation constraint is further designed. The constraint leverages the semantic consistence of the crack regions to promote the network to capture and segment small and complex cracks. Experimental validations on two datasets demonstrate that the proposed approach significantly outperforms state-of-the-art methods, achieving substantial improvements in the fine-grained detail and the continuity of the road crack segmentation.
暂无评论