Xu Wang

Dr. Xu Wang


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 June 2024 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
XWXu Wang
Xu Wang
HYHongwei Yu
Hongwei Yu
XMXiangxin Meng
Xiangxin Meng
HCHongliang Cao
Hongliang Cao
Hongyu ZhangHongyu Zhang
Hongyu Zhang
CHChunming Hu
Chunming Hu
Fault localization (FL) and automated program repair (APR) are two main tasks of automatic software debugging. Compared with traditional methods, deep learning-based approaches have been demonstrated to achieve better performance in FL and APR tasks. However, the existing deep learning-based FL methods ignore the deep semantic features or only consider simple code representations. And for APR tasks, existing template-based APR methods are weak in selecting the correct fix templates for more effective program repair, which are also not able to synthesize patches via the embedded end-to-end code modification knowledge obtained by training models on large-scale bug-fix code pairs. Moreover, in most of FL and APR methods, the model designs and training phases are performed separately, leading to ineffective sharing of updated parameters and extracted knowledge during the training process. This limitation hinders the further improvement in the performance of FL and APR tasks. To solve the above problems, we propose a novel approach called MTL-TRANSFER, which leverages a multi-task learning strategy to extract deep semantic features and transferred knowledge from different perspectives. First, we construct a large-scale open-source bug datasets and implement 11 multi-task learning models for bug detection and patch generation sub-tasks on 11 commonly used bug types, as well as one multi-classifier to learn the relevant semantics for the subsequent fix template selection task. Second, an MLP-based ranking model is leveraged to fuse spectrum-based, mutation-based and semantic-based features to generate a sorted list of suspicious statements. Third, we combine the patches generated by the neural patch generation sub-task from the multi-task learning strategy with the optimized fix template selecting order gained from the multi-classifier mentioned above. Finally, the more accurate FL results, the optimized fix template selecting order, and the expanded patch candidates are combined together to further enhance the overall performance of APR tasks. Our extensive experiments on widely-used benchmark Defects4J show that MTL-TRANSFER outperforms all baselines in FL and APR tasks, proving the effectiveness of our approach. Compared with our previously proposed FL method TRANSFER-FL (which is also the state-of-the-art statement-level FL method), MTL-TRANSFER increases the faults hit by 8/11/12 on Top-1/3/5 metrics (92/159/183 in total). And on APR tasks, the number of successfully repaired bugs of MTL-TRANSFER under the perfect localization setting reaches 75, which is 8 more than our previous APR method TRANSFER-PR. Furthermore, another experiment to simulate the actual repair scenarios shows that MTL-TRANSFER can successfully repair 15 and 9 more bugs (56 in total) compared with TBar and TRANSFER, which demonstrates the effectiveness of the combination of our optimized FL and APR components.

13 September 2024 in Biomedical Signal Processing and Control
Verified
Journal Article
Closed access
QYQingwen Yu
Qingwen Yu
YLYuansen Liu
Yuansen Liu
HLHongyu Li
Hongyu Li
XLXinwen Liu
Xinwen Liu
XBXinlei Bao
Xinlei Bao
XWXu Wang
Xu Wang

20 April 2023 in Future Internet
Verified
Journal Article
Open access
PDF file
XWXu Wang
Xu Wang
Bin ShiBin Shi
Bin Shi
Yili FangYili Fang
Yili Fang
In recent years, the new computing paradigms such as serverless computing, edge computing and blockchain-based computing have attracted much attention in both academia and industrial communities. Distributed systems and applications play fundamental roles in connecting the underlying computers, network and devices for collaboration, as well as providing new services for users. However, due to the increasing complexity of the large-scale and dynamic heterogeneous resource, and the new requirements and features, these distributed systems and applications will face many challenges in terms of their efficiency, flexibility and algorithms. This editorial discusses the state-of-the-art advancements in distributed systems and applications for the emerging computing.

7 October 2026
Verified
Conference Paper
Open access
YSYasai Shi
Yasai Shi
XMXiangxin Meng
Xiangxin Meng
XHXiangjie Huang
Xiangjie Huang
JZJian Zhang
Jian Zhang
TWTianyu Wo
Tianyu Wo
XWXu Wang
Xu Wang

1 October 2026 in Proceedings of the ACM on Software Engineering
Verified
Journal Article
Open access
HYHongwei Yu
Hongwei Yu
XWXu Wang
Xu Wang
JZJian Zhang
Jian Zhang
XMXiangxin Meng
Xiangxin Meng
JLJiarui Li
Jiarui Li
CHChunming Hu
Chunming Hu
Statement-level fault localization (FL) is critical for effective software debugging, as it enables developers to precisely identify faulty lines of code. While traditional spectrum-based, mutation-based, and deep learning-based FL techniques have achieved notable progress, they remain limited in modeling rich fault semantics. Recent advances in large language models (LLMs) offer new opportunities for FL due to their strong capacity for semantic understanding and reasoning over bugs. However, existing LLM-based FL approaches largely treat fault localization as isolated, code-centric prediction, limiting their ability to perform the holistic fault reasoning required for precise statement-level localization. In this paper, we propose FaultScape, a novel LLM-based framework for project-scale, statement-level FL that formulates FL as a multi-view semantic learning and reasoning problem. FaultScape addresses the limitations of existing approaches through three key components. First, we introduce a joint contrastive fine-tuning strategy that trains LLMs on large-scale bug-fix data to explicitly learn fault semantics from multiple complementary views. View-specific fault semantics, including code semantics, fault type, root cause, and repair intent, are learned via supervised binary classification, while cross-view semantic consistency is enforced through contrastive learning. These two objectives are jointly optimized within a unified training loss. The fine-tuned models extract multi-view fault likelihoods as semantic features for each suspicious statement. Second, we adopt a dynamic feature integration module that combines these semantic features with spectrum-based and mutation-based execution features, producing an initial suspiciousness statement ranking. Third, we design an LLM-based, test-guided pairwise re-ranking strategy that explicitly compares highly suspicious candidate statements using failing-test context. By estimating relative fault likelihoods through pairwise comparison rather than independent scoring, the model produces a refined statement-level ranking. We evaluate FaultScape on Defects4J v1.2.0, where it localizes 112/171/196 bugs at Top-1/3/5 out of 395, outperforming state-of-the-art DL-based and LLM-based baselines. On leakage-free benchmarks, FaultScape further localizes 11/18/20 bugs at Top-1/3/5 on ConDefects (31 bugs) and 12/21/22 bugs on GHRB (34 bugs), demonstrating strong generalization to unseen projects. These results show that combining multi-view fault semantics with contrastive, failure-guided reasoning substantially improves the effectiveness and robustness of statement-level fault localization.Statement-level fault localization (FL) is critical for effective software debugging, as it enables developers to precisely identify faulty lines of code. While traditional spectrum-based, mutation-based, and deep learning-based FL techniques have achieved notable progress, they remain limited in modeling rich fault semantics. Recent advances in large language models (LLMs) offer new opportunities for FL due to their strong capacity for semantic understanding and reasoning over bugs. However, existing LLM-based FL approaches largely treat fault localization as isolated, code-centric prediction, limiting their ability to perform the holistic fault reasoning required for precise statement-level localization. In this paper, we propose FaultScape, a novel LLM-based framework for project-scale, statement-level FL that formulates FL as a multi-view semantic learning and reasoning problem. FaultScape addresses the limitations of existing approaches through three key components. First, we introduce a joint contrastive fine-tuning strategy that trains LLMs on large-scale bug-fix data to explicitly learn fault semantics from multiple complementary views. View-specific fault semantics, including code semantics, fault type, root cause, and repair intent, are learned via supervised binary classification, while cross-view semantic consistency is enforced through contrastive learning. These two objectives are jointly optimized within a unified training loss. The fine-tuned models extract multi-view fault likelihoods as semantic features for each suspicious statement. Second, we adopt a dynamic feature integration module that combines these semantic features with spectrum-based and mutation-based execution features, producing an initial suspiciousness statement ranking. Third, we design an LLM-based, test-guided pairwise re-ranking strategy that explicitly compares highly suspicious candidate statements using failing-test context. By estimating relative fault likelihoods through pairwise comparison rather than independent scoring, the model produces a refined statement-level ranking. We evaluate FaultScape on Defects4J v1.2.0, where it localizes 112/171/196 bugs at Top-1/3/5 out of 395, outperforming state-of-the-art DL-based and LLM-based baselines. On leakage-free benchmarks, FaultScape further localizes 11/18/20 bugs at Top-1/3/5 on ConDefects (31 bugs) and 12/21/22 bugs on GHRB (34 bugs), demonstrating strong generalization to unseen projects. These results show that combining multi-view fault semantics with contrastive, failure-guided reasoning substantially improves the effectiveness and robustness of statement-level fault localization.

25 March 2026 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
XWXu Wang
Xu Wang
MCMinghui Chu
Minghui Chu
HYHongwei Yu
Hongwei Yu
CLChuan Luo
Chuan Luo
CHChunming Hu
Chunming Hu
Modern cloud-native architectures demand automated incident identification systems to identify sparse yet critical attribute combinations in multidimensional issue report data. Therefore, there is a need for automated incident detection systems to perform rapid and accurate anomaly detection. Traditional search-based methods are hindered by the combinatorial explosion of attributes and stringent real-time detection requirements, resulting in inefficiency and poor scalability for anomaly identification in large-scale cloud-native systems. To address these challenges, we propose an Adaptive Anomaly Detection for Multi-Dimensional Issue Reports (AADMIR), a combinatorial optimization framework integrating dynamic operator scheduling with scoring-guided neighborhood expansion. Unlike traditional meta-heuristics, AADMIR introduces three core mechanisms specifically designed for high-dimensional issue analysis: (1) Adaptive Operator Selection: A regression-guided dynamic adjustment strategy that automatically selects optimal search operations based on real-time exploration progress, overcoming the ”static strategy trap” in conventional methods. (2) Neighborhood Expansion: A hierarchical second-order neighborhood expansion mechanism addresses the limitations of traditional first-order search by introducing second-order neighborhood. This dynamic topology extends the search scope while maintaining efficiency through probabilistic sampling, effectively balancing global exploration and local exploitation to mitigate local optima entrapment. (3) Regression-Driven Adaptive Judgment: We propose a lightweight, threshold‐based evaluation metric that combines mathematical change‐point detection with information entropy to quantify tuple benefits. By defining an energy metric and using the normalized change threshold, it robustly distinguishes true anomalies from normal fluctuations, avoids false positives inherent in purely entropy‐based methods, and delivers rapid greedy‐step decisions with minimal computational overhead. Experiments on synthetic and real-world datasets demonstrate that AADMIR achieves superior precision, recall, and F1-score compared to state-of-the-art methods, particularly excelling in high-dimensional scenarios.