Nowadays, the pursuit of more empathetic interactions in Emotion AI presents intent detection with increasingly realistic and complex challenges. In this article, we propose a new research task called holistic intent detection (HID), where following five cases are involved: known, unknown, zero–shot, multi-label intents, and intent dependency. Previous studies focus on combinations of no more than three cases. Moreover, when a user conveys multiple intents, existing methods usually only focus on which intents the user expresses, while overlooking the dependencies between intents. To complete the HID task, using a pipeline to concatenate different models is theoretically feasible but impractical due to the low recognition accuracy and error propagation. The powerful reasoning ability shown by large language models (LLMs) makes it possible to complete this task. However, closed-source LLMs exist privacy and cost issues, while there is no effective fine-tuning method when using open-source LLMs with small parameter size for the HID task. Hence, we propose a LLM-based HID framework to recognize holistic intents and their dependency. First, we design a prompt template that considers all cases. Then, different intent selection strategies are used to control the prompt length in the fine-tuning and inference phases to solve the performance degradation caused by lengthy prompts. Experimental results on three datasets demonstrate that our fine-tuned open-source LLM with small parameters outperforms GPT-3.5-turbo by an average of 17.19% in overall accuracy.

Image classification, a fundamental task in computer vision, often suffers from limited labeled samples, a problem commonly referred to as few-shot learning. Model-Agnostic Meta-Learning (MAML) addresses this issue by learning a set of meta-parameters from a variety of similar tasks, enabling rapid adaptation to new tasks within the same distribution using only a few gradient updates. However, traditional MAML frameworks usually encounter two major limitations: one is that they seek a common initialization shared across the entire task distribution, which restricts their capacity to capture the diversity inherent in varying task distributions. The other is that they resort to conventional optimizers (e.g., SGD), which limits their ability to adaptively learn across different domain tasks. To address these limitations, we propose a Task-adaptive Model-Agnostic MetaLearning (TMAML) framework. TMAML improves the existing MAML framework by jointly learning both the initialization and the parameter update rules, tailored for each individual task. The model first learns the domain information of the task as prior parameters, enabling more efficient and effective adaptation in both the initialization and update stages. During initialization, TMAML comprehensively considers the prior parameters of the specific task as well as the intrinsic information of the current samples. When learning the parameter update rules, it fully incorporates the prior parameters of the specific task along with the model's current learning state to guide adaptation more precisely. Experimental results demonstrate that TMAML effectively improves upon existing methods, boosting image classification performance across a wide range of settings. Notably, compared to the state-of-the-art cross-domain few-shot learning method, TMAML achieves an accuracy improvement of up to 3.56 % on 3-domain 5-way 4-shot tasks and up to 5.41 % on 5-domain 5-way 5-shot tasks.

Abstract Smart contracts have significant losses due to various types of vulnerabilities. However, traditional vulnerability detection methods rely extensively on expert rules, resulting in low detection accuracy and poor adaptability to novel attacks. To address these problems, in this paper, deep learning methods are combined with smart contract vulnerability code detection approaches. Abstract syntax trees (ASTs), which are special isomorphic graph structures, are an important bridge between source code and graph neural networks. By learning the AST, the model can understand the semantics of the source code. Moreover, graph neural networks have an increasing ability to address complex heterogeneous graphs. Therefore, control flow graphs are fused with data flow graphs on the basis of the ASTs to build heterogeneous graphs with richer code semantics. Furthermore, multigranularity analysis of the vulnerability detection results is performed, including coarse-grained contract-level vulnerability detection and fine-grained line-level vulnerability detection. Through this multigranularity detection approach, vulnerabilities in contracts can be identified and analysed more comprehensively, providing a richer perspective and more solutions for vulnerability detection. The experimental results show that the proposed multigranularity vulnerability detection method based on heterogeneous graphs (MVD-HG) improves both the accuracy and range of the detected vulnerability types in contract-level vulnerability detection tasks; moreover, in the line-level vulnerability detection task, the MVD-HG model achieves significant results and addresses the shortcomings of existing methods. In addition, based on code generation methods used in related fields, a data enhancement method based on the source code is developed, which effectively expands the experimental dataset to address the reduced credibility of the results due to insufficient amounts of data.

The authentication technology of mobile device users has been studied for decades. To balance security, privacy, and usability, motion sensors-based user authentication methods are widely investigated in recent years. However, existing studies meet the problems such as scarcity of training samples, underutilization of data, poor de-noising ability, insufficient transferability, privacy leakage, and low accuracy. To overcome these difficulties, we propose a system, calledTrapCog, with the following capabilities: 1) In the phase of data collection,TrapCogcan eliminate man-made noise (mislabeling) through differential training based on down-sampling. 2) In the model training stage, the siamese neural network with Long Short-Term Memory (LSTM) as the sub-network is used to achieve sufficient coverage of sample patterns and the transferability of the model. 3) In the phase of real-world authentication, the privacy of the user is tremendously protected through end-side model deployment and local authentication. Experimental results on a dataset composed of 1,513 users with real-world noise show thatTrapCoghas high accuracy and strong transferability, which is much better than state-of-the-art studies.

Advanced Persistent Threat (APT) attacks have caused massive financial loss worldwide. Researchers thereby have proposed a series of solutions to detect APT attacks, such as dynamic/static code analysis, traffic detection, sandbox technology, endpoint detection and response (EDR), etc. However, existing defenses are failed to accurately and effectively defend against the current APT attacks that exhibit strong persistent, stealthy, diverse and dynamic characteristics due to the weak data source integrity, large data processing overhead and poor real-time performance in the process of real-world scenarios. To overcome these difficulties, in this paper we propose APTSHIELD, a stable, efficient and real-time APT detection system for Linux hosts. In the aspect of data collection, audit is selected to stably collect kernel data of the operating system so as to carry out a complete portrait of the attack based on comprehensive analysis and comparison of existing logging tools; In the aspect of data processing, redundant semantics skipping and non-viable node pruning are adopted to reduce the amount of data, so as to reduce the overhead of the detection system; In the aspect of attack detection, an APT attack detection framework based on ATT&CK model is designed to carry out real-time attack response and alarm through the transfer and aggregation of labels. Experimental results on both laboratory and Darpa Engagement show that our system can effectively detect web vulnerability attacks, file-less attacks and remote access trojan attacks, and has a low false positive rate, which adds far more value than the existing frontier work.

Abstract Effective staff scheduling is a critical activity of successful software development management. Due to its difficulty and broad applications in many service delivery scenarios, staff scheduling has been studied for several decades. However, most existing work focus on constructing the working schedules based on a given workforce size. This paper tries to solve a prerequisite issue before performing staff scheduling, i.e., testing whether the already existed manpower can meet the scheduling requirements. Though it is possible to use network flow theory or artificial intelligence (AI) methods like genetic algorithms to solve this problem, their time complexities could be too high to be used for large problem sizes. This paper proposes a constructive method that can derive the minimum staff number for three scheduling problem variants in a linear running time, and in the meantime a corresponding working schedule that can satisfy all the problem constraints can be produced. We not only theoretically show the lower bound for the computation time complexity of our proposed method but also prove its correctness. Moreover, based on the derived minimum staff number, we further explore the genetic algorithm for generating the schedule and compare its performance with our method. The experiments show that our method outperforms the baselines in terms of both effectiveness and efficiency.

In open source communities (e.g., GitHub), developers frequently submit pull requests to fix bugs or add new features during development process. Since the process of pull request is uncoordinated and distributed, it causes massive duplication. Usually, only the first pull request qualified by reviewers can be merged to the main branch of the repository, and the others are regarded as duplication by maintainers. Since the duplication largely aggravates workloads of project reviewers and maintainers, the evolutionary process of open source repositories is delayed. To identify the duplicate pull requests automatically, Ren et al. proposed a state-of-the-art approach that models a pull request by nine features and determine whether a given request is duplicate with the other existing requests or not. Nevertheless, we notice that their approach overlooked the time factor which is a significant feature for the task. In this study, we investigate the influence of time factor and improve the pull request representation. We assume that two pull requests are more likely duplicate when their created time are close to each other. We verify the assumption based on 26 open source repositories from GitHub with over 100,000 pairs of pull requests. We integrate the time feature to the nine features proposed by Ren et al. and the experimental results show that it can substantially improve the performance of Ren et al.'s work by 14.36% and 11.93% in terms of [email protected] and [email protected], respectively.

Given a timed automaton P modeling an implementation and a timed automaton S as a specification, the problem of language inclusion checking is to decide whether the language of P is a subset of that of S. It is known to be undecidable. The problem gets more complicated if non-Zenoness is taken into consideration. A run is Zeno if it permits infinitely many actions within finite time. Otherwise it is non-Zeno. Zeno runs might present in both P and S. It is necessary to check whether a run is Zeno or not so as to avoid presenting Zeno runs as counterexamples of language inclusion checking. In this work, we propose a zone-based semi-algorithm for language inclusion checking with non-Zenoness. It is further improved with simulation reduction based on LU-simulation. Though our approach is not guaranteed to terminate, we show that it does in many cases through empirical study. Our approach has been incorporated into the PAT model checker, and applied to multiple systems to show its usefulness.

Zeno runs, where infinitely many actions occur within finite time, may arise in Timed Automata models. Zeno runs are not feasible in reality and must be pruned during system verification. Thus it is necessary to check whether a run is Zeno or not so as to avoid presenting Zeno runs as counterexamples during model checking. Existing approaches on non-Zenoness checking include either introducing an additional clock in the Timed Automata models or additional accepting states in the zone graphs. In addition, there are approaches proposed for alternative timed modeling languages, which could be generalized to Timed Automata. In this work, we investigate the problem of non-Zenoness checking in the context of model checking LTL properties, not only evaluating and comparing existing approaches but also proposing a new method. To have a systematic evaluation, we develop a software toolkit to support multiple non-Zenoness checking algorithms. The experimental results show the effectiveness of our newly proposed algorithm, and demonstrate the strengths and weaknesses of different approaches.

Non-functional requirements (NFRs) are often regarded as the key success factor in building high quality software. However, most of the requirements elicitation methods are centered on discovering functional requirements only. This paper presents a novel NFRs elicitation approach aiming at empowering requirements analysts with a knowledge repository that aids to the process of capturing precise NFRs during elicitation interviews. The knowledge repository is composed of two layers: the upper layer of feature models and the lower layer of the QoS ontology. The case study of the stock trading domain illustrates the relationships and cooperations of the two layers.