Bihuan Chen

Bihuan Chen

Auto-generated profile, based on publicly available information

Bihuan Chen

Fudan University, China

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
6 February 2024
Verified
Conference Paper
Open access
QGQi Guo
Qi Guo
JCJunming Cao
Junming Cao
XXXiaofei Xie
Xiaofei Xie
SLShangqing Liu
Shangqing Liu
XLXiaohong Li
Xiaohong Li
XPXin Peng
Xin Peng
Code review is an essential activity for ensuring the quality and maintainability of software projects. However, it is a time-consuming and often error-prone task that can significantly impact the development process. Recently, ChatGPT, a cutting-edge language model, has demonstrated impressive performance in various natural language processing tasks, suggesting its potential to automate code review processes. However, it is still unclear how well ChatGPT performs in code review tasks. To fill this gap, in this paper, we conduct the first empirical study to understand the capabilities of ChatGPT in code review tasks, specifically focusing on automated code refinement based on given code reviews. To conduct the study, we select the existing benchmark CodeReview and construct a new code review dataset with high quality. We use CodeReviewer, a state-of-the-art code review tool, as a baseline for comparison with ChatGPT. Our results show that ChatGPT outperforms CodeReviewer in code refinement tasks. Specifically, our results show that ChatGPT achieves higher EM and BLEU scores of 22.78 and 76.44 respectively, while the state-of-the-art method achieves only 15.50 and 62.88 on a high-quality code review dataset. We further identify the root causes for ChatGPT's underperformance and propose several strategies to mitigate these challenges. Our study provides insights into the potential of ChatGPT in automating the code review process, and highlights the potential research directions.

20 February 2026 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
XDXueying Du
Xueying Du
GZGeng Zheng
Geng Zheng
KWKaixin Wang
Kaixin Wang
YZYi Zou
Yi Zou
YWYujia Wang
Yujia Wang
YLYiling Lou
Yiling Lou
Although LLMs have shown promising potential in vulnerability detection, this study reveals their limitations in distinguishing between vulnerable and similar-but-benign patched code (only 0.06 - 0.14 accuracy). It shows that LLMs struggle to capture the root causes of vulnerabilities during vulnerability detection. To address this challenge, we propose enhancing LLMs with multi-dimensional vulnerability knowledge distilled from historical vulnerabilities and fixes. We design a novel knowledge-level Retrieval-Augmented Generation framework Vul-RAG, which improves LLMs with an accuracy increase of 16% - 24% in identifying vulnerable and patched code. Additionally, vulnerability knowledge generated by Vul-RAG can further (1) serve as high-quality explanations to improve manual detection accuracy (from 60% to 77%), and (2) detect 10 previously-unknown bugs in the recent Linux kernel release with 6 assigned CVEs.

14 April 2022 in Empirical Software Engineering
Verified
Journal Article
Closed access
PDF file
KHKaifeng Huang
Kaifeng Huang
BCBihuan Chen
Bihuan Chen
CXCongying Xu
Congying Xu
YWYing Wang
Ying Wang
BSBowen Shi
Bowen Shi
YLYang Liu
Yang Liu

28 April 2025 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
JZJunan Zhang
Junan Zhang
KHKaifeng Huang
Kaifeng Huang
YHYiheng Huang
Yiheng Huang
BCBihuan Chen
Bihuan Chen
RWRuisi Wang
Ruisi Wang
XPXin Peng
Xin Peng
Open source software (OSS) supply chain enlarges the attack surface of a software system, which makes package registries attractive targets for attacks. Recently, multiple package registries have received intensified attacks with malicious packages. Of those package registries, NPM and PyPI are two of the most severe victims. Existing malicious package detectors are developed with features from a list of packages of the same ecosystem and deployed within the same ecosystem exclusively, which is infeasible to utilize the knowledge of a new malicious NPM package detected recently to detect the new malicious package in PyPI. Moreover, existing detectors lack support to model malicious behavior of OSS packages in a sequential way. To address the two limitations, we propose a single detection model using malicious behavior sequence, named Cerebro , to detect malicious packages in NPM and PyPI. We curate a feature set based on a high-level abstraction of malicious behavior to enable multi-lingual knowledge fusing. We organize extracted features into a behavior sequence to model sequential malicious behavior. We fine-tune the pre-trained language model to understand the semantics of malicious behavior. Extensive evaluation has demonstrated the effectiveness of Cerebro over the state-of-the-art as well as the practically acceptable efficiency. Cerebro has detected 683 and 799 new malicious packages in PyPI and NPM, and received 707 thank letters from the official PyPI and NPM teams.

10 October 2022
Verified
Conference Paper
Open access
LZLyuye Zhang
Lyuye Zhang
CLChengwei Liu
Chengwei Liu
ZXZhengzi Xu
Zhengzi Xu
SCSen Chen
Sen Chen
LFLingling Fan
Lingling Fan
YLYang Liu
Yang Liu
To enhance the compatibility in the version control of Java Third-party Libraries (TPLs), Maven adopts Semantic Versioning (SemVer) to standardize the underlying meaning of versions, but users could still confront abnormal execution and crash after upgrades even if compilation and linkage succeed. It is caused by semantic breaking (SemB) issues, such that APIs directly used by users have identical signatures but inconsistent semantics across upgrades. To strengthen compliance with SemVer rules, developers and users should be alerted of such issues. Unfortunately, it is challenging to detect them statically, because semantic changes in the internal methods of APIs are difficult to capture. Dynamic testing can confirmingly uncover some, but it is limited by inadequate coverage. To detect SemB issues over compatible upgrades (Patch and Minor) by SemVer rules, we conduct an empirical study on 180 SemB issues to understand the root causes, inspired by which, we propose Sembid (Semantic Breaking Issue Detector) to statically detect such issues of TPLs for developers and users. Since APIs are directly used by users, Sembid detects and reports SemB issues based on APIs. For a pair of APIs, Sembid walks through the call chains originating from the API to locate breaking changes by measuring semantic diff. Then, Sembid checks if the breaking changes can affect API’s output along call chains. The evaluation showed Sembid achieved recall and precision and outperformed other API checkers on SemB API detection. We also revealed Sembid detected over 3 times more SemB APIs with better coverage than unit tests, the commonly used solution. Furthermore, we carried out an empirical study on 1,629,589 APIs from 546 version pairs of top Java libraries and found there were 2 ∼ 4 times more SemB APIs than those with signature-based issues. Due to various version release strategies, of Patch version pairs and of Minor version pairs had at least one API affected by any breaking.

30 November 2023
Verified
Conference Paper
Open access
KHKaifeng Huang
Kaifeng Huang
BCBihuan Chen
Bihuan Chen
SWSusheng Wu
Susheng Wu
JCJunming Cao
Junming Cao
LMLei Ma
Lei Ma
XPXin Peng
Xin Peng
Deep learning (DL) applications, built upon a heterogeneous and complex DL stack (e.g., Nvidia GPU, Linux, CUDA driver, Python runtime, and TensorFlow), are subject to software and hardware dependencies across the DL stack. One challenge in dependency management across the entire engineering lifecycle is posed by the asynchronous and radical evolution and the complex version constraints among dependencies. Developers may introduce dependency bugs (DBs) in selecting, using and maintaining dependencies. However, the characteristics of DBs in DL stack is still under-investigated, hindering practical solutions to dependency management in DL stack. To bridge this gap, this paper presents the first comprehensive study to characterize symptoms, root causes and fix patterns of DBs across the whole DL stack with 446 DBs collected from StackOverflow posts and GitHub issues. For each DB, we first investigate the symptom as well as the lifecycle stage and dependency where the symptom is exposed. Then, we analyze the root cause as well as the lifecycle stage and dependency where the root cause is introduced. Finally, we explore the fix pattern and the knowledge sources that are used to fix it. Our findings from this study shed light on practical implications on dependency management.