Jinhua Sun

Context: Traditional fuzzing approaches primarily mutate test inputs while relying on manually configured, fixed command-line options. This strategy frequently fails on complex software with massive, interdependent option spaces, leaving much of the program unreachable and reducing testing effectiveness.Objectives: We propose AOFuzz, a sanitizer-guided fuzzer built around adaptive option reconfiguration, which automatically generates command-line option combinations and reconfigures them at run time to reach program locations flagged by sanitizer instrumentation.Methods: AOFuzz couples static code analysis with Large Language Models (LLMs) to generate, validate, and prioritize option combinations against the program's static structure. Using target coverage as its feedback signal, it then reconfigures them. Dynamic scheduling decides when to replace a stagnant combination, and feedback-driven regeneration decides what to generate for sanitizer-flagged target regions that remain uncovered.Results: We evaluated AOFuzz on 17 real-world programs. AOFuzz detects 33.9% more unique bugs than the strongest baseline, the LLM-based ProphetFuzz, and 88.1--116.4% more than FishFuzz++, AFL++, and AFL. In a 24-hour campaign, AOFuzz also uncovered six 0-day vulnerabilities and one regression.Conclusions: Adaptive option reconfiguration turns the command-line option space into a reconfigurable testing dimension, making directed fuzzing practical on option-rich software without manual option tuning.

Action Quality Assessment (AQA)—the task of evaluating how well human actions are performed—is essential in domains such as sports and medicine. Existing AQA methods typically rely on score regression following feature extraction but often neglect the ambiguity inherent in extracted features. In this work, we introduce a novel AQA framework that incorporates a modified attention module to better capture relevant information. Our approach segments video data into clips, extracts features using the I3D network, and applies attention mechanisms to highlight salient features while suppressing irrelevant ones. To assess feature quality, we employ score distribution regression and propose an uncertainty-aware score distribution learning strategy that models features as Gaussian distributions. We further leverage Variational Autoencoders (VAEs) to capture complex latent representations and quantify uncertainty. Extensive experiments on the MTL-AQA and JIGSAWS datasets demonstrate the effectiveness and robustness of our proposed method.

The progress of Internet technology and the portability of mobile communication equipment have provided essential convenience for building an information society. Today, with networked and digital platforms as an important way of socializing, the information carrier is more based on images, videos, etc. The explosive growth of these multimedia data poses a huge challenge to efficient data storage and fast information retrieval. We design an algorithm combining Locality-sensitive Hashing (LSH) and Vision Transformer (ViT) to greatly improve the image matching rate and accuracy, and the algorithm is called ViT-LSH. The principle of LSH is to form the same hash value for points with similar distances in the data set. The results show that the LSH algorithm reduces space consumption and improves query processing efficiency compared with the other five schemes. Therefore, the LSH algorithm can help to quickly get the approximate nearest neighbor query results in a probabilistic manner, thereby ensuring the accuracy of the query results and the efficiency of time and space. The ViT divides the image into fixed-size image blocks ( Patch), uses Linear projection to project the Patch flatten to the specified dimension, gets the Tokens sequence, and adopts the sequence as the input of the feature to realize a new segmentation mode. The ViT does not employ down-sampling so that the image resolution is not reduced. The modeling of global information is a new semantic segmentation mode. The steps of ViT-LSH are as follows: first, preprocess the image, such as denoising, then segment the image, then extract features by ViT, and then reduce features' dimension. The LSH performs hash coding on the image features. After calculating the image's hash code, the image's similarity can be directly calculated by the Hamming distance, sorted according to the similarity, and the matching results are obtained. The experimental results exhibit that ViT-LSH can greatly enhance the efficiency and accuracy of image matching.

To better monitor the usage status of electric energy for production and living devices, the paper has put forward an electric energy monitoring system based on Modbus protocol, starting by analyzing the demand for monitoring and management of users’ electric energy. A three-layer structure has been used in the system, which consists of the sensing layer, control layer, and application layer. The lower computer of the sensing layer used the one-master-multi-slave mode to collect the power data of multiple devices in the bus in real time and transmits the voltage, current, power consumption, and other data of power-consumption devices to the host through the RS-485 interface, Modbus protocol, and GPRS network. Meanwhile, the lower computer interacts with the remote user monitoring terminal to realize the monitoring and management of power consumption data of the devices. The test has shown that the system can steadily monitor and manage the power consumption status of the devices in real time, and can be applied to energy saving, monitoring of power-consumption devices, property management, and other fields, which have good application prospects.