Pun memes, which combine wordplay with visual elements, represent a popular form of humor in Chinese online communications.Despite their prevalence, current Vision-Language Models (VLMs) lack systematic evaluation in understanding and applying these culturallyspecific multimodal expressions.In this paper, we introduce PUNMEMECN, a novel benchmark designed to assess VLMs' capabilities in processing Chinese pun memes across three progressive tasks: pun meme detection, pun meme sentiment analysis, and chat-driven meme response.PUNMEMECN consists of 1,959 Chinese memes (653 pun memes and 1,306 non-pun memes) with comprehensive annotations of punchlines, sentiments, and explanations, alongside 2,008 multi-turn chat conversations incorporating these memes.Our experiments indicate that state-of-the-art VLMs struggle with Chinese pun memes, particularly with homophone wordplay, even with Chainof-Thought prompting.Notably, punchlines in memes can effectively conceal potentially harmful content from AI detection.These findings underscore the challenges in cross-cultural multimodal understanding and highlight the need for culture-specific approaches to humor comprehension in AI systems.

The architecture of a neural network serves as the foundation for deep learning models and plays a crucial role in efficient training and application. Designing high-performance neural network architectures typically requires extensive expertise, along with numerous trials and optimizations of hyperparameters and network structure parameters. This process demands significant time and effort. To automate the optimization of LSTM (Long Short-Term Memory), this paper combines evolutionary algorithms for LSTM neural network architecture search and hyperparameter tuning. Additionally, to prevent LSTM from getting stuck in local optima during backpropagation, particle swarm optimization is employed to initialize the weights of LSTM. Experimental results on real-world datasets demonstrate that the proposed algorithm can provide different neural network architectures tailored to different datasets, achieving competitive predictive capabilities.

Active Learning is an essential method for label-efficient deep learning. As a Bayesian active learning method, Bayesian Active Learning by Disagreement (BALD) successfully selects the most representative samples by maximizing the mutual information between the model prediction and model parameters. However, when applied to a batch acquisition mode, like batch construction with greedy search, BALD suffers from poor performance, especially with noises of near-duplicate data. To address this shortcoming, we propose a diverse beam search optimized batch active learning method, which explores a graph for every batch construction by expanding the highest-scored samples of a predetermined number. To avoid near duplicate beam branches (very similar beams generated from the same root and similar samples), which is undesirable for lacking diverse representations in the feature space, we design a self-adapted constraint within candidate beams. The proposed method is able to acquire data that can better represent the distribution of the unlabeled pool, and at the same time, be significantly different from existing beams. We observe that the proposed method achieves higher batch performance than the baseline methods on three benchmark datasets.

The tremendous advancements in deep learning techniques have empowered question answering(QA) systems with the capability of dealing with various tasks. Many commercial QA systems, such as Siri, Google Home, and Alexa, have been deployed to assist people in different daily activities. However, modern QA systems are often designed to deal with different topics and task formats, which makes both the test collection and labeling tasks difficult and thus threats their quality.

Nowadays, the explosive growth of time series data and the idea of automatically classifying them brought chances of advanced analysis and machine cognitive processes in various domains. Hundreds of Time Series Classification (TSC) algorithms have been proposed during the last decade. Most of them need the large quantity of labeled training data for achieving good precision. However, we noticed that under most of the training scenarios, the large-scale supervised training datasets are not readily available. We thus proposed a few shot deep learning neural network framework PIN-BA (Prototypical Inception Network with Cross Branch Attention) for time series classification with only limited training data. We use CNN (Convolutional Neural Network) with branches of different reception windows for capturing the features of different time window scales. We design cross branch attention schemes based on prototypical networks to emphasize the crucial features' information during classification. A few experiments were conducted on the famous UCR Time Series Data Sets. Experimental results demonstrate that our proposed framework outperforms the other TSC models, such like 1NN-DTW and InceptionTime under the few shot training data scenarios.

This work describes an encoder pre-training procedure using frame-wise label to improve the training of streaming recurrent neural network transducer (RNN-T) model.Streaming RNN-T trained from scratch usually performs worse than nonstreaming RNN-T.Although it is common to address this issue through pre-training components of RNN-T with other criteria or frame-wise alignment guidance, the alignment is not easily available in end-to-end manner.In this work, frame-wise alignment, used to pre-train streaming RNN-T's encoder, is generated without using a HMM-based system.Therefore an allneural framework equipping HMM-free encoder pre-training is constructed.This is achieved by expanding the spikes of CTC model to their left/right blank frames, and two expanding strategies are proposed.To our best knowledge, this is the first work to simulate HMM-based frame-wise label using CTC model for pre-training.Experiments conducted on LibriSpeech and MLS English tasks show the proposed pre-training procedure, compared with random initialization, reduces the WER by relatively 5%∼11% and the emission latency by 60 ms.Besides, the method is lexicon-free, so it is friendly to new languages without manually designed lexicon.

Log parsers first convert large-scale and unstructured system logs into structured data, and then cluster them into groups for anomaly detection and monitoring. However, the security vulnerabilities of the log parsers have not been unveiled yet. In this paper, to our best knowledge, we take the first step to propose a novel real-time black-box attack framework LogBug in which attackers slightly modify the logs to deviate the analysis result (i.e., evading the anomaly detection) without knowing the learning model and parameters of the log parser. We have empirically evaluated LogBug on five emerging log parsers using system logs collected from five different systems. The results demonstrate that LogBug can greatly reduce the accuracy of log parsers with minor perturbations in real time.

Nowadays, explosive growth of ontologies are used for managing data in various domains. They usually own different vocabularies and structures following different fashions. Ontology alignment finding semantic correspondences between elements of these ontologies can effectively facilitate the data communication and novel application creation in many practical scenarios. However, we noticed that, the traditional parametric ontology mapping methods still depend on individualistic abilities for setting proper parameters for mapping. When trying to utilize artificial neural networks for the automatic ontology mapping, the training data are found insufficient in most of the cases. This paper analyzes these problems, and proposes a few-shot ontology alignment model, which can automatically learn how to map two ontologies from only a few training links between their element pairs. The proposed model applies the Siamese neural network in computer vision on ontology alignment and designs an attention detection network learning the attention weights for different ontology attributes. A few experiments that conducted on the anatomy ontology alignment show that our model achieves good performance (94.3% of F-measure) with 200 training alignments without traditional parametric setting.

Today, most of the IoT platforms including the standardizations for them are encountering a new stage called "Semantic Interoperability" which is expected to promise the intensive analytics and high level cross-domain intelligence in future. Although multiple IoT platforms such as oneM2M and OPC UA developed the outline architectures for IoT semantic implementation, the large cost on manual semantic annotation, ontology construction and metadata transferring hinder the practical use of these architectures. Efficient data aggression and computing perception for automatic semantic annotation and ontology learning become urgent for solving this problem. This paper did an exploratory investigation into these present researches and the future of semantic-enabled IoT. Real time stream reasoning technology and hierarchical clustering technology for automatic semantic annotation and reducing the semantic meta-data traffic are analyzed. The potential ways these two kinds of technologies can be combined and the possible benefits are estimated. A software architecture for facilitating these technologies in IoT is designed with introducing the respectively analyzed related algorithms and methodologies.

Collaborative Web search improves search quality by users' working in cooperation and is a subset of social search. Current Web browsers and search engines provide limited support for it. However, it is easier for experts, who are familiar with some topics, to fulfill they needs through search engines due to their backgrounds, domain knowledge and so on. A sharing experts' experiences approach should be struck based on today's Web browsers and major search engines. This paper presents a convenient way for users to share and utilize experts' experiences through a Web browser toolbar for collaborative Web search. The toolbar an catch search histories and favorites and display recommendations for every user in a popular Web browser through integrating with mainstream search engines like Google, Yahoo!, et al. These collected users' data are uploaded to a recommendation server, in which recommendations are built according to some rules based on an utilizing experts' experiences approach. The toolbar can download some valuable recommendations merging into default search list for prompting a searcher. The core of our proposed approach is a scalable method to measure "to what degree a user is an expert" for a given topic and to detect an expert's experiences based on a hierarchical user profile. Experiments showed that the novel collaborative Web search way is acceptant to users and experts' experiences improved search quality when compared to standard Google rankings. More importantly, results verified our hypothesis that a significant improvement on search quality can be achieved by utilizing experts' experiences.