Haoyu Wang

Hongzhou Rao, Yanjie Zhao, Xinyi Hou et al.

The rapid advancement of large language models (LLMs) has redefined artificial intelligence (AI), pushing the boundaries of AI research and enabling unbounded possibilities for both academia and the industry. However, LLM development faces increasingly complex challenges throughout its lifecycle, yet no existing research systematically explores these challenges and solutions from the perspective of software engineering (SE) approaches. To fill the gap, we systematically analyze research status throughout the LLM development lifecycle, divided into six phases: requirements engineering, dataset construction, model development and enhancement, testing and evaluation, deployment and operations, and maintenance and evolution. We then conclude by identifying the key challenges for each phase and presenting potential research directions to address these challenges. In general, we provide valuable insights from an SE perspective to facilitate future advances in LLM development.

Yu Liang, Arin Chaudhuri, Haoyu Wang

Dimension reduction and visualization of high-dimensional data have become very important research topics because of the rapid growth of large databases in data science. In this paper, we propose using a generalized sigmoid function to model the distance similarity in both high- and low-dimensional spaces. In particular, the parameter b is introduced to the generalized sigmoid function in low-dimensional space, so that we can adjust the heaviness of the function tail by changing the value of b. Using both simulated and real-world data sets, we show that our proposed method can generate visualization results comparable to those of uniform manifold approximation and projection (UMAP), which is a newly developed manifold learning technique with fast running speed, better global structure, and scalability to massive data sets. In addition, according to the purpose of the study and the data structure, we can decrease or increase the value of b to either reveal the finer cluster structure of the data or maintain the neighborhood continuity of the embedding for better visualization. Finally, we use domain knowledge to demonstrate that the finer subclusters revealed with small values of b are meaningful.

Liu Wang, Ren He, Haoyu Wang et al.

As the COVID-19 pandemic emerged in early 2020, a number of malicious actors have started capitalizing the topic. Although a few media reports mentioned the existence of coronavirus-themed mobile malware, the research community lacks the understanding of the landscape of the coronavirus-themed mobile malware. In this paper, we present the first systematic study of coronavirus-themed Android malware. We first make efforts to create a daily growing COVID-19 themed mobile app dataset, which contains 4,322 COVID-19 themed apk samples (2,500 unique apps) and 611 potential malware samples (370 unique malicious apps) by the time of mid-November, 2020. We then present an analysis of them from multiple perspectives including trends and statistics, installation methods, malicious behaviors and malicious actors behind them. We observe that the COVID-19 themed apps as well as malicious ones began to flourish almost as soon as the pandemic broke out worldwide. Most malicious apps are camouflaged as benign apps using the same app identifiers (e.g., app name, package name and app icon). Their main purposes are either stealing users' private information or making profit by using tricks like phishing and extortion. Furthermore, only a quarter of the COVID-19 malware creators are habitual developers who have been active for a long time, while 75% of them are newcomers in this pandemic. The malicious developers are mainly located in US, mostly targeting countries including English-speaking countries, China, Arabic countries and Europe. To facilitate future research, we have publicly released all the well-labelled COVID-19 themed apps (and malware) to the research community. Till now, over 30 research institutes around the world have requested our dataset for COVID-19 themed research.

Ningyu He, Lei Wu, Haoyu Wang et al.

In this paper, we present the first large-scale and systematic study to characterize the code reuse practice in the Ethereum smart contract ecosystem. We first performed a detailed similarity comparison study on a dataset of 10 million contracts we had harvested, and then we further conducted a qualitative analysis to characterize the diversity of the ecosystem, understand the correlation between code reuse and vulnerabilities, and detect the plagiarist DApps. Our analysis revealed that over 96% of the contracts had duplicates, while a large number of them were similar, which suggests that the ecosystem is highly homogeneous. Our results also suggested that roughly 9.7% of the similar contract pairs have exactly the same vulnerabilities, which we assume were introduced by code clones. In addition, we identified 41 DApps clusters, involving 73 plagiarized DApps which had caused huge financial loss to the original creators, accounting for 1/3 of the original market volume.

Haoyu Wang, Yawen Guan, Brian J Reich

Kriging is the predominant method used for spatial prediction, but relies on the assumption that predictions are linear combinations of the observations. Kriging often also relies on additional assumptions such as normality and stationarity. We propose a more flexible spatial prediction method based on the Nearest-Neighbor Neural Network (4N) process that embeds deep learning into a geostatistical model. We show that the 4N process is a valid stochastic process and propose a series of new ways to construct features to be used as inputs to the deep learning model based on neighboring information. Our model framework outperforms some existing state-of-art geostatistical modelling methods for simulated non-Gaussian data and is applied to a massive forestry dataset.

Arin Chaudhuri, Carol Sadek, Deovrat Kakde et al.

Support vector data description (SVDD) is a popular anomaly detection technique. The SVDD classifier partitions the whole data space into an inlier region, which consists of the region near the training data, and an outlier region, which consists of points away from the training data. The computation of the SVDD classifier requires a kernel function, for which the Gaussian kernel is a common choice. The Gaussian kernel has a bandwidth parameter, and it is important to set the value of this parameter correctly for good results. A small bandwidth leads to overfitting such that the resulting SVDD classifier overestimates the number of anomalies, whereas a large bandwidth leads to underfitting and an inability to detect many anomalies. In this paper, we present a new unsupervised method for selecting the Gaussian kernel bandwidth. Our method exploits a low-rank representation of the kernel matrix to suggest a kernel bandwidth value. Our new technique is competitive with the current state of the art for low-dimensional data and performs extremely well for many classes of high-dimensional data. Because the mathematical formulation of SVDD is identical with the mathematical formulation of one-class support vector machines (OCSVM) when the Gaussian kernel is used, our method is equally applicable to Gaussian kernel bandwidth tuning for OCSVM.

Hansi Jiang, Haoyu Wang, Wenhao Hu et al.

Support vector data description (SVDD) is a machine learning technique that is used for single-class classification and outlier detection. The idea of SVDD is to find a set of support vectors that defines a boundary around data. When dealing with online or large data, existing batch SVDD methods have to be rerun in each iteration. We propose an incremental learning algorithm for SVDD that uses the Gaussian kernel. This algorithm builds on the observation that all support vectors on the boundary have the same distance to the center of sphere in a higher-dimensional feature space as mapped by the Gaussian kernel function. Each iteration involves only the existing support vectors and the new data point. Moreover, the algorithm is based solely on matrix manipulations; the support vectors and their corresponding Lagrange multiplier $α_i$'s are automatically selected and determined in each iteration. It can be seen that the complexity of our algorithm in each iteration is only $O(k^2)$, where $k$ is the number of support vectors. Experimental results on some real data sets indicate that FISVDD demonstrates significant gains in efficiency with almost no loss in either outlier detection accuracy or objective function value.