Adversarial attacks for natural language processing aim to perturb input text and fool victim models. Adversarial triggers, which are a new form of adversarial perturbation used for textual adversarial attacks, have gained research attention in recent years. Existing approaches of adversarial trigger attacks typically attempt to improve the semantics and fluency of triggers. However, they are input-agnostic and fail to consider varied contexts. In this paper, we propose a novel kind of adversarial triggers, Contextual Adversarial Triggers (CAT). We design a three-step framework that trains a masked language model as a generator to produce contextual adversarial triggers. CAT, which is a context-aware and varied adversarial trigger, is enabled to attack victim models effectively for exposing the vulnerabilities of these models. Adversarial training on adversarial examples with CAT can enhance the model's robustness against general adversarial trigger attacks. Experimental results show that CAT outperforms existing adversarial triggers in revealing the vulnerabilities of models, and the model's robustness is improved when CAT is used for adversarial training. Furthermore, experimental results demonstrate that CAT has significant advantages in both attack effectiveness and efficiency compared with typical adversarial attack methods.

Recent works have proposed methods of generating synthetic data automatically for unsupervised Grammatical Error Correction (GEC). Although a large amount of synthetic data is generated at a low cost, it is unrealistic and of poor quality. The copying phenomenon of synthetic data prevents GEC models from learning the semantic knowledge of contextual language. In this paper, we design an instruction format and use the masking strategy in both an erroneous sentence and the corresponding instruction consistently to alleviate the impact of the copy phenomenon. We also propose a novel approach, InstructGEC, which integrates the knowledge of grammatical detection into GEC models with instruction tuning to address the low-quality issue. Experiments are conducted on English and Chinese GEC datasets and results demonstrate that our method outperforms state-of-the-art unsupervised GEC methods.

Detecting rumors on social media has become a critical task in combating misinformation.Existing propagation-based rumor detection methods often focus on the static propagation graph, overlooking that rumor propagation is inherently dynamic and incremental in the real world.Recently propagation-based rumor detection models attempt to use the dynamic graph that is associated with coarse-grained temporal information.However, these methods fail to capture the long-term time dependency and detailed temporal features of propagation.To address these issues, we propose a novel adaptive Sliding Window and memory-augmented Attention Model (SWAM) for rumor detection.The adaptive sliding window divides the sequence of posts into consecutive disjoint windows based on the propagation rate of nodes.We also propose a memory-augmented attention to capture the long-term dependency and the depth of nodes in the propagation graph.Multi-head attention mechanism is applied between nodes in the memorybank and incremental nodes to iteratively update the memorybank, and the depth information of nodes is also considered.Finally, the propagation features of nodes in the memorybank are utilized for rumor detection.Experimental results on two public real-world datasets demonstrate the effectiveness of our model compared with the state-of-the-art baselines.

With the rapid development of the Internet and the widespread use of social media, the proliferation of multimodal misinformation combining images and text poses serious risks to societal trust, individual well-being, and the integrity of AI models trained on such data. Recently, the automatic detection multimodal misinformation has become an essential area of research. However, traditional methods often rely on hierarchical neural networks that compress and fuse modalities, potentially overlooking deeper interactions between modalities and reducing model interpretability. In this paper, we present a novel Multimodal Taylor Series (MTS) network for detecting multimodal misinformation. The MTS network leverages Taylor series expansion to explicitly capture both low-order and high-order interactions between modalities, which also enhances interpretability by decomposing the model's processing into distinct terms. Additionally, the proposed MTS network avoids exponential parameter growth and maintains linear scalability, allowing the model to effectively capture complex cross-modal correlations. Extensive experiments on three benchmark datasets demonstrate that the MTS network significantly outperforms state-of-the-art models. We have open-sourced the code and logs at: https://github.com/OneForAllSama/MTS.

Grammatical Error Correction (GEC) is usually considered as a translation task where an erroneous sentence is treated as the source language and the corrected sentence as the target language. The state-of-the-art GEC models often adopt transformer-based sequence-to-sequence architecture of machine translation. However, most of these approaches ignore the syntactic information because the syntax of an erroneous sentence is also full of errors and not beneficial to GEC. In this paper, we propose a novel Error-Correction Constituent Parsing (ECCP) task which uses the constituent parsing of corrected sentences to avoid the harmful effect of the erroneous sentence. We also propose an architecture that includes one encoder and two decoders. There are millions of parameters in transformer-based GEC models, and the labeled training data is substantially less than synthetic pre-training data. Therefore, adapter layers are added to the proposed architecture, and adapter tuning is used for fine-tuning our model to alleviate the low-resource issue. We conduct experiments on CoNLL-2014, BEA-2019, and JFLEG test datasets in unsupervised and supervised settings. Experimental results show that our method outperforms the-state-of-art baselines and achieves superior performance on all datasets.

Predicate abstraction techniques have been shown to be a powerful technique for verifying imperative programs, which can solve the problem of state space explosion pretty well. Among them, lazy abstraction with interpolation-based refinement also called the IMPACT approach has gained increasing popularity in the last years. However, despite its high efficiency, the IMPACT fails to work out some kinds of the programs because the interpolants produced by interpolant solver are so bad to make the verification divergent. According to the features of some of these programs, we extend the IMPACT method to make it applicable for them. In addition to its basic ones, two other operations are introduced to the IMPACT refinement to guide it produce reasonal interpolants which are helpful for the verification process to converge. The experiments on the benchmark of SV-COMP2020 show the potential of the extended approach.

The k-induction algorithm is a well-known verification technique that combines bounded model checking with an inductive approach to verify program safety. Many software verification tools have implemented it. The performance of k-induction is dependent on the inductive loop invariant. However, it is hard to generate useful invariants, especially when the loop construction is complex and nested by other constructions such as branches. To solve this problem, we introduce value analysis into k-induction, which is used to generate a set of custom loop invariants according to safety properties. In contrast to general loop invariants provided by loop invariant generators, custom invariants are made to verify the property, and the induction method converges with them more quickly. Our experiments show that combining k-induction with the invariants were produced by value analysis significantly increases effectiveness and efficiency and outperforms existing implementations of k-induction based software verification in terms of successful verification results.

The explosive growth of fake news on social media has drawn great concern both from industrial and academic communities. There has been an increasing demand for fake news detection due to its detrimental effects. Generally, news content is condensed and full of knowledge entities. However, existing methods usually focus on the textual contents and social context, and ignore the knowledge-level relationships among news entities. To address this limitation, in this paper, we propose a novel Knowledge-aware Attention Network (KAN) that incorporates external knowledge from knowledge graph for fake news detection. Firstly, we identify entity mentions in news contents and align them with the entities in knowledge graph. Then, the entities and their contexts are used as external knowledge to provide complementary information. Finally, we design News towards Entities (N-E) attention and News towards Entities and Entity Contexts (N-E^2C) attention to measure the importances of knowledge. Thus, our proposed model can incorporate both semantic-level and knowledge-level representations of news to detect fake news. Experimental results on three public datasets show that our model outperforms the state-of-the-art methods, and also validate the effectiveness of knowledge attention.

Program verification techniques have gained increasing popularity in academic and industrial circles during the last years. Predicate abstraction is a traditional and practical verification technique, which can solve the problem of state space explosion pretty well. Many software verification tools have implemented it. But these implementations are not user-friendly, or scalable. Aimed at these problems, we describe and implement a new automatic predicate abstraction framework, CChecker, for proving the safety of procedural programs with integer assignments. CChecker is a whole system composed of two parts: front and back end. The front end preprocesses and parses the source programs into logic models based on Clang. And the back end resolves the models based on Z3 to get software safety property. At last, the experiments show the potential of CChecker.

A key problem in any automatic software verification system is the inference of loop invariants. When analyzing program structures involving disjunctive semantics, abstract interpretation has the problem of precision loss. Thus, some techniques were proposed to decompose such loop structures into a semantically equivalent sequence of loops with conjunctive semantics whose invariants can be generated by abstract interpretation directly. However, these works assumed that the iteration processes of nested branches are separate without consideration of non-monotone loop structures where those interweave with each other. In order to solve this problem, we present a novel static analysis technique for non-monotone loops. It analyzes loop convergence condition and traces the transfer between nested branches of finite non-monotone loops. With analytical results, it generates the loop invariants with precise semantics. Meanwhile, it takes advantage of cyclical nature of result expressions to restrict search space and accelerate computation procedure. Finally, experimental results show the potential of our approach, which is also helpful for reasoning about certain program security properties.

With the increasing popularity of mobile Internet, it can bring great business value for Telecommunication (Telco) operators to provide users with better services in a timely manner. Understanding the change of mobile user's activity can be a great help for operators to increase user experience and avoid the user churn. In this paper, we predict short-term user activity with massive Mobile Broadband (MBB) data. We conduct experiments with a large scale and real-world dataset of Telco operators, which includes MBB data of more than three million users. The experimental results show that gradient boosting decision tree is the effective model for the prediction. In addition, we show that the user activity is highly correlated with individual features. Features, which are associated with personal daily habits, tend to make people active the next day. In contract, features, which happen for specific purpose and are less related to the individual habits, can make people inactive the next day.

With the universal application of software cluster systems, their reliability is drawing more and more attention from academia to industry. A cluster system is a kind of software load-sharing system (LSS) whose reliability is significantly dependent on system software. Therefore, traditional reliability analysis methods for hardware LSSs are not applicable for cluster systems. In this paper, we develop a reliability analysis model for redundant cluster systems consisting of initial servers and cold standby servers used to replace failed ones. System reliability process is modeled with a state-based non-homogeneous Markov process (NHMH), where each state corresponds to a non-homogeneous Poisson processe (NHPP). NHPP arrival rate is expressed using Cox's proportional hazard model (PHM) in terms of cumulative and instantaneous workload of system software. In addition to redundant cluster systems without repair, the model also can be extended to analyze those with restart. The analysis results are meaningful to support cluster management and design decisions. Finally, the evaluation experiments show the potential of our model.