Deductive reasoning is a crucial logical capability that assists us in solving complex problems based on existing knowledge. Although augmented by Chain-of-Thought prompts, Large Language Models (LLMs) might not follow the correct reasoning paths. Enhancing the deductive reasoning abilities of LLMs, and leveraging their extensive built-in knowledge for various reasoning tasks, remains an open question. Attempting to mimic the human deductive reasoning paradigm, we propose a multi-stage Syllogistic-Reasoning Framework of Thought (SR-FoT) that enables LLMs to perform syllogistic deductive reasoning to handle complex knowledge-based reasoning tasks. Our SR-FoT begins by interpreting the question and then uses the interpretation and the original question to propose a suitable major premise. It proceeds by generating and answering minor premise questions in two stages to match the minor premises. Finally, it guides LLMs to use the previously generated major and minor premises to perform syllogistic deductive reasoning to derive the answer to the original question. Extensive and thorough experiments on knowledge-based reasoning tasks have demonstrated the effectiveness and advantages of our SR-FoT.

CNN-based methods have made great progress in single-image rain removal. Most recent methods improve performance by increasing the depth of the network. To fully extract local and global features while reducing inference time, we propose a top-to-down attribute-insensitive multiscale hourglass network for rain streak and raindrop removal. For the rain removal task, we expect that the constructed network can accurately identify the various attributes of the rain information characteristics of the small target. Considering the difference in the size, shape, direction and density of rain streak and raindrop, inspired by the performance of hourglass architecture to capture multiscale features in human pose estimation, we introduce an attribute-insensitive hourglass module to recognize the attributes of rain streak and raindrop in a unified framework. This feature extraction module could capture the characteristics of rain streak and raindrop with different attributes. This stacked hourglass blocks down-sample features and then up-samples them back to the original resolution based on discrete wavelet transform and inverse discrete wavelet transform. We perform extensive experiments on five synthetic and real-world de-raining datasets to validate the effectiveness of our proposed network on rain streak and raindrop removal. The qualitative and quantitative results show that our method is suitable for removing rain streak and raindrop in a unified framework. We present the results of generalization and ablation study for key components, we also report the accuracy of semantic segmentation after preprocessing with all rain removal methods. Our source code will be available on the GitHub: https://github.com/Ruini94/AIMHNet .

Data in the visual world often present long-tailed distributions. However, learning high-quality representations and classifiers for imbalanced data is still challenging for data-driven deep learning models. In this work, we aim at improving the feature extractor and classifier for long-tailed recognition via contrastive pretraining and feature normalization, respectively. First, we carefully study the influence of contrastive pretraining under different conditions, showing that current self-supervised pretraining for long-tailed learning is still suboptimal in both performance and speed. We thus propose a new balanced contrastive loss and a fast contrastive initialization scheme to improve previous long-tailed pretraining. Second, based on the motivative analysis on the normalization for classifier, we propose a novel generalized normalization classifier that consists of generalized normalization and grouped learnable scaling. It outperforms traditional inner product classifier as well as cosine classifier. Both the two components proposed can improve recognition ability on tail classes without the expense of head classes. We finally build a unified framework that achieves competitive performance compared with state of the arts on several long-tailed recognition benchmarks and maintains high efficiency.

The accuracy of scene text recognition has been significantly improved, which can be attributed to the development of deep learning. However, the blurring and low-resolution text images usually lead to unsatisfactory results in text recognition. Several researchers design super-resolution models that adopt convolutional neural networks (CNNs) to relieve the image blurring, while these models are limited to the receptive field of the convolution kernel and fail to extract the long-distance semantic relations of text images enough. In this paper, we propose a CNN-Transformer Text Super Resolution Network (CTTSR) to capture the semantic features of text images by the multi-head attention mechanism of the transformer. Furthermore, we propose the text position loss to optimize the network and make the text regions of images more effectively detectable. Experimental results demonstrate that our model can improve the quality of images and outperform the existing methods in text recognition tasks.

Over the years, population protocols with the goal of reaching consensus have been studied in great depth. However, many systems in the real-world do not result in all agents eventually reaching consensus, but rather in the opposite: they converge to a state of rich diversity. Consider for example task allocation in ants. If eventually all ants perform the same task, then the colony will perish (lack of food, no brood care, etc.). Then, it is vital for the survival of the colony to have a diverse set of tasks and enough ants working on each task. What complicates matters is that ants need to switch tasks periodically to adjust the needs of the colony; e.g., when too many foragers fell victim to other ant colonies. A further difficulty is that not all tasks are equally important and maybe they need to keep certain proportions in the distribution of the task. How can ants keep a healthy and balanced allocation of tasks?

Given a graph G of n vertices, where each vertex is initially attached an opinion of either red or blue. We investigate a random process known as the Best-of-three voting. In this process, at each time step, every vertex chooses three neighbours at random and adopts the majority colour. We study this process for a class of graphs with minimum degree d = nα, where α = Ømega((łog łog n)-1). We prove that if initially each vertex is red with probability greater than $1/2+δ, and blue otherwise, where δ ≥ (łog d)-C for some C>0, then with high probability this dynamic reaches a final state where all vertices are red within O(łog łog n) + O(łog(δ-1)) steps.

Over the past few years, with the trend of mobile computing, Mobile Ad hoc Network (MANET) has become one of the most important wireless communication mechanisms among all. Unlike traditional network, MANET does not have a fixed infrastructure, every single node in the network works as both a receiver and a transmitter. Nodes directly communicate with each other when they are both within their communication ranges. Otherwise, they rely on their neighbors to store and forward packets. As MANET does not require any fixed infrastructure and it is capable of self configuring, these unique characteristics made MANET ideal to be deployed in a remote or mission critical area like military use or remote exploration. However, the open medium and wide distribution of nodes in MANET leave it vulnerable to various means of attacks. It is crucial to develop suitable intrusion detection scheme to protect MANET from malicious attackers. In our previous research, we have proposed a mechanism called Enhanced Adaptive Acknowledgement (EAACK) scheme. Nevertheless, it suffers from the threat that it fails to detect misbehaving node when the attackers are smart enough to forge the acknowledgement packets. In this paper, we introduce Digital Signature Algorithm (DSA) into the EAACK scheme, and investigate the performance of DSA in MANET. The purpose of this paper is to present an improved version of EAACK called EAACK2 that performs better in the presence of false misbehavior and partial dropping.

There has been a tremendous growth in the use of wireless communication in the past few decades. Mobile Ad hoc NETwork (MANET) is one of the most important one among various wireless communication mechanisms. In MANET, each node in a network performs as both a transmitter and a receiver. They rely on each other to store and forward packets. Its unique infrastructureless network and self-configuring capability makes it ideal for many mission critical applications, including military use and remote exploration. However, these characteristics also make MANET vulnerable to passive and active attacks due to its open medium, changing topology and lack of centralized monitoring. To address the new security challenges, Intrusion Detection System (IDS) is required to detect the malicious attackers before they can accomplish any significant damages to the network. Many existing IDSs for MANETs are based upon Watchdog mechanism. In this paper, we propose a new IDS called Enhanced Adaptive ACKnowledgement (EAACK) that solves four significant problems of Watchdog mechanism, which are ambiguous collisions, receiver collisions, limited transmission power and false misbehavior report. We use Network Simulator 2 to simulate the proposed mechanism and compare the results with existing mechanisms.

Nowadays, wireless sensor networks are deployed in a wide range of applications such as military. To enable sinks to avoid physical attacks from adversaries, most of WSNs adopt sink-location privacy mechanisms. By utilizing these mechanisms, an adversary cannot analyze packet traffic and perform hop-by-hop trace-back, and thus deduce the location of a sink. In this paper, we propose an attack approach to track anonymous sinks. It utilizes a Pseudo-Noise (PN) code to mark a data flow in an invisible manner. An adversary is able to interfere with a source nodepsilas traffic by embedding a secure signal into the nodepsilas traffic. The signal is carried along with the traffic from the source node to the sink. Therefore, the attacker can recognize the location of a sink node by tracking the invisible secure signal. Through our simulation experiments, we conclude that the proposed attack approach is able to track an anonymous sink without additional traffic overhead.