Beiqi Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Beiqi Zhang has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Peng Liang
School of Computer Science, Wuhan University, Wuhan, China
8 shared publications - Chong Wang
School of Computer Science Wuhan University Wuhan China
3 shared publications - Xiyu Zhou
School of Computer Science, Wuhan University, China
3 shared publications - Zengyang Li
School of Computer Science, Central China Normal University, WuHan, HuBei, China
3 shared publications - Aakash Ahmad
Lancaster University
2 shared publications - David Lo
School of Computing and Information Systems, Singapore Management University, Singapore, Singapore
2 shared publications
Latest Publications
- Open Access
- 3 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
A Comprehensive Evaluation of Parameter-Efficient Fine-Tuning on Code Smell Detection
Code smells are indicators of suboptimal design that negatively impact software quality. However, automated code smell detection remains a persistent challenge: heuristics-based tools suffer from high sensitivity to threshold selection and inherent subjectivity, while Machine Learning (ML) and Deep Learning (DL) models yield unsatisfactory performance. Although Large Language Models (LLMs) offer a promising solution to these limitations, their adaptation is impeded by the prohibitive cost of full fine-tuning and the lack of “LM-ready” benchmarks, as existing code smell datasets largely rely on software metrics or contain noisy, unverified labels. To address these gaps and enable a rigorous assessment of efficient LLM adaptation for code smell detection, we present a comprehensive study comprising two synergistic contributions. First, we constructed a high-quality, source-code-centric benchmark covering four non-trivial code smells— Complex Conditional , Complex Method , Feature Envy , and Data Class —and ensured its reliability through a meticulous two-stage manual review. Second, leveraging this benchmark, we systematically evaluated the effectiveness of four Parameter-Efficient Fine-Tuning (PEFT) methods (i.e., prompt tuning, prefix tuning, LoRA, and \((IA)^{3}\) ) across four Small LMs (SLMs) and five LLMs. We benchmarked these PEFT-tuned LMs against a broad suite of baselines, including traditional heuristics-based detectors ( DesigniteJava and PMD ), DL-based approaches ( DeepSmells and AE-Dense ), and cutting-edge general-purpose LLMs (GPT-4o-mini and DeepSeek-v3) with In-Context Learning. Our results demonstrate that PEFT methods match or exceed full fine-tuning performance while significantly reducing peak GPU memory usage, and outperform all baselines with MCC improvements ranging from 0.33% to 13.69%, particularly for some code smells. Furthermore, we provide actionable insights into the choice of PEFT method for code smell detection, which should be guided by the model used, data availability, and computational resources. Our replication package is available at [1] to support reproducibility and future extensions.- 3 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
Open AccessA Comprehensive Evaluation of Parameter-Efficient Fine-Tuning on Code Smell Detection
Code smells are indicators of suboptimal design that negatively impact software quality. However, automated code smell detection remains a persistent challenge: heuristics-based tools suffer from high sensitivity to threshold selection and inherent subjectivity, while Machine Learning (ML) and Deep Learning (DL) models yield unsatisfactory performance. Although Large Language Models (LLMs) offer a promising solution to these limitations, their adaptation is impeded by the prohibitive cost of full fine-tuning and the lack of “LM-ready” benchmarks, as existing code smell datasets largely rely on software metrics or contain noisy, unverified labels. To address these gaps and enable a rigorous assessment of efficient LLM adaptation for code smell detection, we present a comprehensive study comprising two synergistic contributions. First, we constructed a high-quality, source-code-centric benchmark covering four non-trivial code smells— Complex Conditional , Complex Method , Feature Envy , and Data Class —and ensured its reliability through a meticulous two-stage manual review. Second, leveraging this benchmark, we systematically evaluated the effectiveness of four Parameter-Efficient Fine-Tuning (PEFT) methods (i.e., prompt tuning, prefix tuning, LoRA, and \((IA)^{3}\) ) across four Small LMs (SLMs) and five LLMs. We benchmarked these PEFT-tuned LMs against a broad suite of baselines, including traditional heuristics-based detectors ( DesigniteJava and PMD ), DL-based approaches ( DeepSmells and AE-Dense ), and cutting-edge general-purpose LLMs (GPT-4o-mini and DeepSeek-v3) with In-Context Learning. Our results demonstrate that PEFT methods match or exceed full fine-tuning performance while significantly reducing peak GPU memory usage, and outperform all baselines with MCC improvements ranging from 0.33% to 13.69%, particularly for some code smells. Furthermore, we provide actionable insights into the choice of PEFT method for code smell detection, which should be guided by the model used, data availability, and computational resources. Our replication package is available at [1] to support reproducibility and future extensions. - Open Access
- 15 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
- Vol. 35 (8) , 1-38
Using LLMs in Generating Design Rationale for Software Architecture Decisions
Design Rationale (DR) for software architecture decisions refers to the reasoning underlying architectural choices, which provides valuable insights into the different phases of the architecting process throughout software development. However, in practice, DR is often inadequately documented due to a lack of motivation and effort from developers. With the recent advancements in Large Language Models (LLMs), their capabilities in text comprehension, reasoning, and generation may enable the generation and recovery of DR for architecture decisions. In this study, we evaluated the performance of LLMs in generating DR for architecture decisions. First, we collected 50 Stack Overflow (SO) posts, 25 GitHub issues, and 25 GitHub discussions related to architecture decisions to construct a dataset of 100 architecture-related problems. Then, we selected five LLMs to generate DR for the architecture decisions with three prompting strategies, including zero-shot, chain of thought (CoT), and LLM-based agents. With the DR provided by human experts as ground truth, the Precision of LLM-generated DR with the three prompting strategies ranges from 0.267 to 0.278, Recall from 0.627 to 0.715, and F1-score from 0.351 to 0.389. Besides, 64.45% to 69.42% of the arguments of DR not mentioned by human experts are also helpful, 4.12% to 4.87% of the arguments have uncertain correctness, and 1.59% to 3.24% of the arguments are potentially misleading. To further understand the trustworthiness and applicability of LLM-generated DR in practice, we conducted semi-structured interviews with six practitioners. Based on the experimental and interview results, we discussed the pros and cons of the three prompting strategies, the strengths and limitations of LLM-generated DR, and the implications for the practical use of LLM-generated DR.
- 15 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
- Vol. 35 (8) , 1-38
Open AccessUsing LLMs in Generating Design Rationale for Software Architecture Decisions
Design Rationale (DR) for software architecture decisions refers to the reasoning underlying architectural choices, which provides valuable insights into the different phases of the architecting process throughout software development. However, in practice, DR is often inadequately documented due to a lack of motivation and effort from developers. With the recent advancements in Large Language Models (LLMs), their capabilities in text comprehension, reasoning, and generation may enable the generation and recovery of DR for architecture decisions. In this study, we evaluated the performance of LLMs in generating DR for architecture decisions. First, we collected 50 Stack Overflow (SO) posts, 25 GitHub issues, and 25 GitHub discussions related to architecture decisions to construct a dataset of 100 architecture-related problems. Then, we selected five LLMs to generate DR for the architecture decisions with three prompting strategies, including zero-shot, chain of thought (CoT), and LLM-based agents. With the DR provided by human experts as ground truth, the Precision of LLM-generated DR with the three prompting strategies ranges from 0.267 to 0.278, Recall from 0.627 to 0.715, and F1-score from 0.351 to 0.389. Besides, 64.45% to 69.42% of the arguments of DR not mentioned by human experts are also helpful, 4.12% to 4.87% of the arguments have uncertain correctness, and 1.59% to 3.24% of the arguments are potentially misleading. To further understand the trustworthiness and applicability of LLM-generated DR in practice, we conducted semi-structured interviews with six practitioners. Based on the experimental and interview results, we discussed the pros and cons of the three prompting strategies, the strengths and limitations of LLM-generated DR, and the implications for the practical use of LLM-generated DR.
- Open Access
LessLeak-Bench: A First Investigation of Data Leakage in LLMs Across 83 Software Engineering Benchmarks
Large Language Models (LLMs) are widely utilized in software engineering (SE) tasks, such as code generation and automated program repair. However, their reliance on extensive and often undisclosed pre-training datasets raises significant concerns about data leakage, where the evaluation benchmark data is unintentionally ``seen'' by LLMs during the model's construction phase. The data leakage issue could largely undermine the validity of LLM-based research and evaluations. Despite the increasing use of LLMs in the SE community, there is no comprehensive study that assesses the extent of data leakage in SE benchmarks for LLMs yet. To address this gap, this paper presents the first large-scale analysis of data leakage in 83 SE benchmarks concerning LLMs. Our results show that in general, data leakage in SE benchmarks is minimal, with average leakage ratios of only 4.8\%, 2.8\%, and 0.7\% for Python, Java, and C/C++ benchmarks, respectively. However, some benchmarks exhibit relatively higher leakage ratios, which raises concerns about their bias in evaluation. For instance, QuixBugs and BigCloneBench have leakage ratios of 100.0\% and 55.7\%, respectively. Furthermore, we observe that data leakage has a substantial impact on LLM evaluation. We also identify key causes of high data leakage, such as the direct inclusion of benchmark data in pre-training datasets and the use of coding platforms like LeetCode for benchmark construction. To address the data leakage, we introduce \textbf{LessLeak-Bench}, a new benchmark that removes leaked samples from the 83 SE benchmarks, enabling more reliable LLM evaluations in future research. Our study enhances the understanding of data leakage in SE benchmarks and provides valuable insights for future research involving LLMs in SE.
Open AccessLessLeak-Bench: A First Investigation of Data Leakage in LLMs Across 83 Software Engineering Benchmarks
Large Language Models (LLMs) are widely utilized in software engineering (SE) tasks, such as code generation and automated program repair. However, their reliance on extensive and often undisclosed pre-training datasets raises significant concerns about data leakage, where the evaluation benchmark data is unintentionally ``seen'' by LLMs during the model's construction phase. The data leakage issue could largely undermine the validity of LLM-based research and evaluations. Despite the increasing use of LLMs in the SE community, there is no comprehensive study that assesses the extent of data leakage in SE benchmarks for LLMs yet. To address this gap, this paper presents the first large-scale analysis of data leakage in 83 SE benchmarks concerning LLMs. Our results show that in general, data leakage in SE benchmarks is minimal, with average leakage ratios of only 4.8\%, 2.8\%, and 0.7\% for Python, Java, and C/C++ benchmarks, respectively. However, some benchmarks exhibit relatively higher leakage ratios, which raises concerns about their bias in evaluation. For instance, QuixBugs and BigCloneBench have leakage ratios of 100.0\% and 55.7\%, respectively. Furthermore, we observe that data leakage has a substantial impact on LLM evaluation. We also identify key causes of high data leakage, such as the direct inclusion of benchmark data in pre-training datasets and the use of coding platforms like LeetCode for benchmark construction. To address the data leakage, we introduce \textbf{LessLeak-Bench}, a new benchmark that removes leaked samples from the 83 SE benchmarks, enabling more reliable LLM evaluations in future research. Our study enhances the understanding of data leakage in SE benchmarks and provides valuable insights for future research involving LLMs in SE.
- Open Access
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 2230-2234
Copilot-in-the-Loop: Fixing Code Smells in Copilot-Generated Python Code using Copilot
As one of the most popular dynamic languages, Python experiences a decrease in readability and maintainability when code smells are present. Recent advancements in Large Language Models have sparked growing interest in AI-enabled tools for both code generation and refactoring. GitHub Copilot is one such tool that has gained widespread usage. Copilot Chat, released in September 2023, functions as an interactive tool aimed at facilitating natural language-powered coding. However, limited attention has been given to understanding code smells in Copilot-generated Python code and Copilot Chat's ability to fix the code smells. To this end, we built a dataset comprising 102 code smells in Copilot-generated Python code. Our aim is to first explore the occurrence of code smells in Copilot-generated Python code and then evaluate the effectiveness of Copilot Chat in fixing these code smells employing different prompts. The results show that 8 out of 10 types of code smells can be detected in Copilot-generated Python code, among which Multiply-Nested Container is the most common one. For these code smells, Copilot Chat achieves a highest fixing rate of 87.1%, showing promise in fixing Python code smells generated by Copilot itself. In addition, the effectiveness of Copilot Chat in fixing these smells can be improved by providing more detailed prompts.
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 2230-2234
Open AccessCopilot-in-the-Loop: Fixing Code Smells in Copilot-Generated Python Code using Copilot
As one of the most popular dynamic languages, Python experiences a decrease in readability and maintainability when code smells are present. Recent advancements in Large Language Models have sparked growing interest in AI-enabled tools for both code generation and refactoring. GitHub Copilot is one such tool that has gained widespread usage. Copilot Chat, released in September 2023, functions as an interactive tool aimed at facilitating natural language-powered coding. However, limited attention has been given to understanding code smells in Copilot-generated Python code and Copilot Chat's ability to fix the code smells. To this end, we built a dataset comprising 102 code smells in Copilot-generated Python code. Our aim is to first explore the occurrence of code smells in Copilot-generated Python code and then evaluate the effectiveness of Copilot Chat in fixing these code smells employing different prompts. The results show that 8 out of 10 types of code smells can be detected in Copilot-generated Python code, among which Multiply-Nested Container is the most common one. For these code smells, Copilot Chat achieves a highest fixing rate of 87.1%, showing promise in fixing Python code smells generated by Copilot itself. In addition, the effectiveness of Copilot Chat in fixing these smells can be improved by providing more detailed prompts.
- 3 June 2024
- Published by Springer Nature in Empirical Software Engineering
- Vol. 29 (4) , 1-45
Demystifying code snippets in code reviews: a study of the OpenStack and Qt communities and a practitioner survey
- 3 June 2024
- Published by Springer Nature in Empirical Software Engineering
- Vol. 29 (4) , 1-45
Demystifying code snippets in code reviews: a study of the OpenStack and Qt communities and a practitioner survey
- Open Access
Demystifying Code Snippets in Code Reviews: A Study of the OpenStack and Qt Communities and A Practitioner Survey
Code review is widely known as one of the best practices for software quality assurance in software development. In a typical code review process, reviewers check the code committed by developers to ensure the quality of the code, during which reviewers and developers would communicate with each other in review comments to exchange necessary information. As a result, understanding the information in review comments is a prerequisite for reviewers and developers to conduct an effective code review. Code snippet, as a special form of code, can be used to convey necessary information in code reviews. For example, reviewers can use code snippets to make suggestions or elaborate their ideas to meet developers' information needs in code reviews. However, little research has focused on the practices of providing code snippets in code reviews. To bridge this gap, we conduct a mixed-methods study to mine information and knowledge related to code snippets in code reviews, which can help practitioners and researchers get a better understanding about using code snippets in code review. Specifically, our study includes two phases: mining code review data and conducting practitioners' survey. The study results highlight that reviewers can provide code snippets in appropriate scenarios to meet developers' specific information needs in code reviews, which will facilitate and accelerate the code review process.
Open AccessDemystifying Code Snippets in Code Reviews: A Study of the OpenStack and Qt Communities and A Practitioner Survey
Code review is widely known as one of the best practices for software quality assurance in software development. In a typical code review process, reviewers check the code committed by developers to ensure the quality of the code, during which reviewers and developers would communicate with each other in review comments to exchange necessary information. As a result, understanding the information in review comments is a prerequisite for reviewers and developers to conduct an effective code review. Code snippet, as a special form of code, can be used to convey necessary information in code reviews. For example, reviewers can use code snippets to make suggestions or elaborate their ideas to meet developers' information needs in code reviews. However, little research has focused on the practices of providing code snippets in code reviews. To bridge this gap, we conduct a mixed-methods study to mine information and knowledge related to code snippets in code reviews, which can help practitioners and researchers get a better understanding about using code snippets in code review. Specifically, our study includes two phases: mining code review data and conducting practitioners' survey. The study results highlight that reviewers can provide code snippets in appropriate scenarios to meet developers' specific information needs in code reviews, which will facilitate and accelerate the code review process.
- 17 October 2023
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 33 (11n12) , 1653-1672
Demystifying Practices, Challenges and Expected Features of Using GitHub Copilot
With the advances in machine learning, there is a growing interest in AI-enabled tools for autocompleting source code. GitHub Copilot, also referred to as the “AI Pair Programmer”, has been trained on billions of lines of open source GitHub code, and is one of such tools that has been increasingly used since its launch in June 2021. However, little effort has been devoted to understanding the practices, challenges, and expected features of using Copilot in programming for auto-completed source code from the point of view of practitioners. To this end, we conducted an empirical study by collecting and analyzing the data from Stack Overflow (SO) and GitHub Discussions. More specifically, we searched and manually collected 303 SO posts and 927 GitHub discussions related to the usage of Copilot. We identified the programming languages, Integrated Development Environments (IDEs), technologies used with Copilot, functions implemented, benefits, limitations, and challenges when using Copilot. The results show that when practitioners use Copilot: (1) The major programming languages used with Copilot are JavaScript and Python, (2) the main IDE used with Copilot is Visual Studio Code, (3) the most common used technology with Copilot is Node.js, (4) the leading function implemented by Copilot is data processing, (5) the main purpose of users using Copilot is to help generate code, (6) the significant benefit of using Copilot is useful code generation, (7) the main limitation encountered by practitioners when using Copilot is difficulty of integration, and (8) the most common expected feature is that Copilot can be integrated with more IDEs. Our results suggest that using Copilot is like a double-edged sword, which requires developers to carefully consider various aspects when deciding whether or not to use it. Our study provides empirically grounded foundations that could inform software developers and practitioners, as well as provide a basis for future investigations on the role of Copilot as an AI pair programmer in software development.
- 17 October 2023
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 33 (11n12) , 1653-1672
Demystifying Practices, Challenges and Expected Features of Using GitHub Copilot
With the advances in machine learning, there is a growing interest in AI-enabled tools for autocompleting source code. GitHub Copilot, also referred to as the “AI Pair Programmer”, has been trained on billions of lines of open source GitHub code, and is one of such tools that has been increasingly used since its launch in June 2021. However, little effort has been devoted to understanding the practices, challenges, and expected features of using Copilot in programming for auto-completed source code from the point of view of practitioners. To this end, we conducted an empirical study by collecting and analyzing the data from Stack Overflow (SO) and GitHub Discussions. More specifically, we searched and manually collected 303 SO posts and 927 GitHub discussions related to the usage of Copilot. We identified the programming languages, Integrated Development Environments (IDEs), technologies used with Copilot, functions implemented, benefits, limitations, and challenges when using Copilot. The results show that when practitioners use Copilot: (1) The major programming languages used with Copilot are JavaScript and Python, (2) the main IDE used with Copilot is Visual Studio Code, (3) the most common used technology with Copilot is Node.js, (4) the leading function implemented by Copilot is data processing, (5) the main purpose of users using Copilot is to help generate code, (6) the significant benefit of using Copilot is useful code generation, (7) the main limitation encountered by practitioners when using Copilot is difficulty of integration, and (8) the most common expected feature is that Copilot can be integrated with more IDEs. Our results suggest that using Copilot is like a double-edged sword, which requires developers to carefully consider various aspects when deciding whether or not to use it. Our study provides empirically grounded foundations that could inform software developers and practitioners, as well as provide a basis for future investigations on the role of Copilot as an AI pair programmer in software development.
- Open Access
Demystifying Practices, Challenges and Expected Features of Using GitHub Copilot
With the advances in machine learning, there is a growing interest in AI-enabled tools for autocompleting source code. GitHub Copilot has been trained on billions of lines of open source GitHub code, and is one of such tools that has been increasingly used since its launch in June 2021. However, little effort has been devoted to understanding the practices, challenges, and expected features of using Copilot in programming for auto-completed source code from the point of view of practitioners. To this end, we conducted an empirical study by collecting and analyzing the data from Stack Overflow (SO) and GitHub Discussions. We searched and manually collected 303 SO posts and 927 GitHub discussions related to the usage of Copilot. We identified the programming languages, Integrated Development Environments (IDEs), technologies used with Copilot, functions implemented, benefits, limitations, and challenges when using Copilot. The results show that when practitioners use Copilot: (1) The major programming languages used with Copilot are JavaScript and Python, (2) the main IDE used with Copilot is Visual Studio Code, (3) the most common used technology with Copilot is Node.js, (4) the leading function implemented by Copilot is data processing, (5) the main purpose of users using Copilot is to help generate code, (6) the significant benefit of using Copilot is useful code generation, (7) the main limitation encountered by practitioners when using Copilot is difficulty of integration, and (8) the most common expected feature is that Copilot can be integrated with more IDEs. Our results suggest that using Copilot is like a double-edged sword, which requires developers to carefully consider various aspects when deciding whether or not to use it. Our study provides empirically grounded foundations that could inform developers and practitioners, as well as provide a basis for future investigations.
Open AccessDemystifying Practices, Challenges and Expected Features of Using GitHub Copilot
With the advances in machine learning, there is a growing interest in AI-enabled tools for autocompleting source code. GitHub Copilot has been trained on billions of lines of open source GitHub code, and is one of such tools that has been increasingly used since its launch in June 2021. However, little effort has been devoted to understanding the practices, challenges, and expected features of using Copilot in programming for auto-completed source code from the point of view of practitioners. To this end, we conducted an empirical study by collecting and analyzing the data from Stack Overflow (SO) and GitHub Discussions. We searched and manually collected 303 SO posts and 927 GitHub discussions related to the usage of Copilot. We identified the programming languages, Integrated Development Environments (IDEs), technologies used with Copilot, functions implemented, benefits, limitations, and challenges when using Copilot. The results show that when practitioners use Copilot: (1) The major programming languages used with Copilot are JavaScript and Python, (2) the main IDE used with Copilot is Visual Studio Code, (3) the most common used technology with Copilot is Node.js, (4) the leading function implemented by Copilot is data processing, (5) the main purpose of users using Copilot is to help generate code, (6) the significant benefit of using Copilot is useful code generation, (7) the main limitation encountered by practitioners when using Copilot is difficulty of integration, and (8) the most common expected feature is that Copilot can be integrated with more IDEs. Our results suggest that using Copilot is like a double-edged sword, which requires developers to carefully consider various aspects when deciding whether or not to use it. Our study provides empirically grounded foundations that could inform developers and practitioners, as well as provide a basis for future investigations.
- Open Access
RoseMatcher: Identifying the Impact of User Reviews on App Updates
$\textbf{Context}$: The release planning of mobile apps has become an area of active research, with most studies centering on app analysis through release notes in the Apple App Store and tracking user reviews via issue trackers. However, the correlation between these release notes and user reviews in App Store remains understudied. $\textbf{Objective}$: In this paper, we introduce $\textit{RoseMatcher}$, a novel automatic approach to match relevant user reviews with app release notes and identify matched pairs with high confidence. $\textbf{Methods}$: We collected 944 release notes and 1,046,862 user reviews from 5 mobile apps in the Apple App Store as research data to evaluate the effectiveness and accuracy of $\textit{RoseMatcher}$, and conducted deep content analysis on matched pairs. $\textbf{Results}$: Our evaluation shows that $\textit{RoseMatcher}$ can reach a hit ratio of 0.718 for identifying relevant matched pairs, and with the manual labeling and content analysis of 984 relevant pairs, we identify 8 roles that user reviews play in app updates according to the relationship between release notes and user reviews in the relevant matched pairs. $\textbf{Conclusions}$: Our findings indicate that both app development teams and users pay close attention to release notes and user reviews, with release notes typically addressing feature requests, bug reports, and complaints, and user reviews offering positive, negative, and constructive feedback. Overall, the study highlights the importance of the communication between app development teams and users in the release planning of mobile apps, with relevant reviews tending to be posed within a short period before and after the release of release notes, with the average time interval between the post time of release notes and user reviews being approximately one year.Open AccessRoseMatcher: Identifying the Impact of User Reviews on App Updates
$\textbf{Context}$: The release planning of mobile apps has become an area of active research, with most studies centering on app analysis through release notes in the Apple App Store and tracking user reviews via issue trackers. However, the correlation between these release notes and user reviews in App Store remains understudied. $\textbf{Objective}$: In this paper, we introduce $\textit{RoseMatcher}$, a novel automatic approach to match relevant user reviews with app release notes and identify matched pairs with high confidence. $\textbf{Methods}$: We collected 944 release notes and 1,046,862 user reviews from 5 mobile apps in the Apple App Store as research data to evaluate the effectiveness and accuracy of $\textit{RoseMatcher}$, and conducted deep content analysis on matched pairs. $\textbf{Results}$: Our evaluation shows that $\textit{RoseMatcher}$ can reach a hit ratio of 0.718 for identifying relevant matched pairs, and with the manual labeling and content analysis of 984 relevant pairs, we identify 8 roles that user reviews play in app updates according to the relationship between release notes and user reviews in the relevant matched pairs. $\textbf{Conclusions}$: Our findings indicate that both app development teams and users pay close attention to release notes and user reviews, with release notes typically addressing feature requests, bug reports, and complaints, and user reviews offering positive, negative, and constructive feedback. Overall, the study highlights the importance of the communication between app development teams and users in the release planning of mobile apps, with relevant reviews tending to be posed within a short period before and after the release of release notes, with the average time interval between the post time of release notes and user reviews being approximately one year.