Cross-Modal Consistency with Aesthetic Similarity for Multimodal False Information Detection

With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to multimodalinformation exchange and fusion, with many methods attempting to integrate unimodal features to generatemultimodal news representations. However, they still need to fully explore the hierarchical and complex semanticcorrelations between different modal contents, severely limiting their performance detecting multimodal falseinformation. This work proposes a two-stage detection framework for multimodal false information detection,called ASMFD, which is based on image aesthetic similarity to segment and explores the consistency andinconsistency features of images and texts. Specifically, we first use the Contrastive Language-Image Pre-training(CLIP) model to learn the relationship between text and images through label awareness and train an imageaesthetic attribute scorer using an aesthetic attribute dataset. Then, we calculate the aesthetic similarity betweenthe image and related images and use this similarity as a threshold to divide the multimodal correlation matrixinto consistency and inconsistencymatrices. Finally, the fusionmodule is designed to identify essential features fordetectingmultimodal false information. In extensive experiments on four datasets, the performance of the ASMFDis superior to state-of-the-art baseline methods.

False Negative Sample Detection for Graph Contrastive Learning

Recently,self-supervised learning has shown great potential in Graph Neural Networks (GNNs) through contrastive learning,which aims to learn discriminative features for each node without label information. The key to graph contrastive learning is data augmentation. The anchor node regards its augmented samples as positive samples,and the rest of the samples are regarded as negative samples,some of which may be positive samples. We call these mislabeled samples as “false negative” samples,which will seriously affect the final learning effect. Since such semantically similar samples are ubiquitous in the graph,the problem of false negative samples is very significant. To address this issue,the paper proposes a novel model,False negative sample Detection for Graph Contrastive Learning (FD4GCL),which uses attribute and structure-aware to detect false negative samples. Experimental results on seven datasets show that FD4GCL outperforms the state-of-the-art baselines and even exceeds several supervised methods.

A Hybrid Method for False Data Injection Attack Detection in Smart Grid Based on Variational Mode Decomposition and OS-ELM

Accurate state estimation is critical to wide-area situational awareness of smart grid.However,recent research found that power system state estimators are vulnerable to a new type of cyber-attack,called false data injection attack(FDIA).In order to ensure the security of power system operation and control,a hybrid FDIA detection mechanism utilizing temporal correlation is proposed.The proposed mechanism combines Variational Mode Decomposition(VMD)technology and machine learning.For the purpose of identifying the features of FDIA,VMD is used to decompose the system state time series into an ensemble of components with different frequencies.Furthermore,due to the lack of online model updating ability in a traditional extreme learning machine,an OS-extreme learning machine(OSELM)which has sequential learning ability is used as a detector for identifying FDIA.The proposed detection mechanism is evaluated on the IEEE-14 bus system using real load data from an independent system operator in New York.Apart from detection accuracy,the impact of attack intensity and environment noise on the performance of the proposed method are tested.The simulation results demonstrate the efficiency and robustness of our method.

Adaptive coherence estimator based on the Krylov subspace technique for airborne radar

A novel adaptive detector for airborne radar space-time adaptive detection （STAD） in partially homogeneous environments is proposed. The novel detector combines the numerically stable Krylov subspace technique and diagonal loading technique, and it uses the framework of the adaptive coherence estimator （ACE）. It can effectively detect a target with low sample support. Compared with its natural competitors, the novel detector has higher proba- bility of detection （PD）, especially when the number of the training data is low. Moreover, it is shown to be practically constant false alarm rate （CFAR）.

Combination of differential discrimination and direct discrimination in pulsed laser timeof-flight systems

The output amplitude of the differential circuit is studied for differential discrimination in pulsed laser time-offlight systems. Based on the studies of the probability of detection and the probability of false alarms, the minimum detectable input signal of differential discrimination can be calculated. The results indicate that the differential discrimination detectability of the small signal will be reduced. A combined discrimination is proposed in this Letter to improve the time resolution of the large signal and ensure the probability of detection of the small signal at the same time. A proper value of the circuit parameter is found to balance the time resolutions of the small and large signals.