arXiv subjects

Peng Liang

Publications and source records attributed to Peng Liang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Rule Taxonomy and Evolution in AI IDEs: A Mining and Survey Study

The adoption of AI-powered Integrated Development Environments (AI IDEs) has introduced "Rules" as a novel software artifact, allowing developers to persistently inject project-specific constraints and architectural guidelines into the context of Large Language Models (LLMs). Despite their role in aligning AI behavior with developer intent, the taxonomy, evolution, and practical impact of these rules remain largely unexplored. To bridge this gap, we conducted a mixed-methods empirical study on AI IDE rules. By mining 83 open-source projects and extracting 7,310 rules, we established a comprehensive taxonomy comprising 5 primary and 25 secondary categories. We then triangulated these artifacts with survey responses from 99 practitioners. Our analysis identified a contrast between developer priorities and actual configurations: while practitioners rate architectural constraints as highly important, rule files in repositories primarily consist of low-level workflow and code formatting constraints. Furthermore, our analysis of 1,540 rule evolution events revealed that rules are updated frequently. Repository data further indicate that rule evolution is primarily driven by constructive context expansions (29.17%) and enrichments (26.59%). In contrast, surveyed developers reported modifying rules primarily to correct AI errors (77.78%), typically by adding new negative constraints rather than editing existing ones. Finally, an artifact compliance assessment of 160 rule evolution events revealed that updating rules significantly improves the adherence of software artifacts, with the average artifact compliance rate increasing by 22.99% (from 49.14% to 72.13%) following an update. Our study provides empirical insights that can help developers optimize prompting strategies and guide tool builders in designing automated conflict-detection and context-management mechanisms for AI IDEs.

cs.SE↗

arXiv2026

Bridging Requirements and Architecture: Multi-Agent Orchestration with External Knowledge and Hierarchical Memory

Software architecture design is a critical yet inherently complex and knowledge-intensive phase that requires balancing competing quality attributes and adapting to evolving requirements. Traditionally, this process has been time-consuming, labor-intensive, and heavily reliant on architects, often resulting in limited exploration of alternative architectural decompositions and styles, especially under the pressures of agile development. While LLM-based agents have shown promising performance across various software engineering tasks, their application to architecture design remains relatively scarce and requires systematic exploration. To address these challenges, we proposed MAAD (Multi-Agent Architecture Design), a knowledge-driven framework that orchestrates four specialized agents (i.e., Analyst, Modeler, Designer and Evaluator) to autonomously and collaboratively transform requirements specifications into comprehensive, multi-view architectural blueprints with quality attribute assessments. MAAD incorporates RAG to inject recognized architectural standards and patterns into the workflow and leverages a hierarchical memory mechanism that captures design history for iterative refinement. We evaluated MAAD through comparative experiments against MetaGPT, using quantitative architecture-level metrics across 10 case studies and qualitative feedback from industry architects on 10 real-world specifications. Results show that MAAD generates more complete, modular, and traceable architectures than the baseline, and its dedicated Evaluator agent autonomously produces structured quality evaluation reports that significantly reduce manual validation efforts. Furthermore, we found that the quality of the generated architecture heavily depends on the underlying LLM's reasoning capacity, with GPT-5.2 and Qwen3.5 outperforming other models across most evaluation settings.

cs.SE↗

arXiv2026

DRReduce: Enhancing Syntax-Guided Program Reduction with Dependency Reconstruction

Program reduction is a technique for simplifying large, failure-inducing programs into minimal reproducible test cases. Language-specific tools such as CReduce achieve strong performance by leveraging deep semantic knowledge of C/C++, but are tightly coupled to a single language family. Language-agnostic reducers such as Perses address this by applying syntax-guided search across any grammar, yet share a fundamental limitation: deleting a node or subtree in isolation often breaks semantic coherence causing the property checker to reject the deletion and forcing the reducer to backtrack, limiting overall reduction effectiveness and efficiency. In this paper, we propose DRReduce, a framework that bridges this gap by augmenting language-agnostic syntactic reduction with a lightweight semantic layer: dependency reconstruction, which repairs program dependencies broken by a deletion in order to preserve the semantic validity of intermediate programs and increase the acceptance rate of the property checker. DRReduce constructs a semantic dependency graph from the input program, performs semantically coherent deletions with dependency reconstruction, and delegates further minimization to a syntax-guided reducer. We implement DRReduce for C and Java and evaluate it on real-world bug-triggering programs. Compared to SOTA syntax-guided reducers, DRReduce achieves average size reductions of 51.9%, 14.9%, and 19.8% over Perses, WDD, and CDD respectively, while completing reduction faster on the majority of programs. Compared to language-specific tools, DRReduce achieves results comparable to CReduce and Latra without any language-specific transformation rules, at 3.3x and 1.2x higher efficiency than CReduce and Latra on average, respectively. An ablation study confirms that dependency reconstruction reduces query invocations by 80.2%, reduction time by 58.7%, and final token count by over 55.1%.

cs.SE↗

arXiv2026

An Insight into Security Code Review with LLMs: Capabilities, Obstacles, and Influential Factors

Security code review is a time-consuming and labor-intensive process typically requiring integration with automated security defect detection tools. However, existing security analysis tools struggle with poor generalization, high false positive rates, and coarse detection granularity. Large Language Models (LLMs) have been considered promising candidates for addressing those challenges. In this study, we conducted an empirical study to explore the potential of LLMs in detecting security defects during code review. Specifically, we evaluated the performance of seven LLMs under five different prompts and compared them with state-of-the-art static analysis tools. We also performed linguistic and regression analyses for the two top-performing LLMs to identify quality problems in their responses and factors influencing their performance. Our findings show that: (1) In security code review, LLMs significantly outperform state-of-the-art static analysis tools, and the reasoning-optimized LLM performs better than general-purpose LLMs. (2) DeepSeek-R1 achieves the highest performance, followed by GPT-4 provided in the ChatGPT platform. The optimal prompt for DeepSeek-R1 incorporates both the commit message and chain-of-thought (CoT) guidance, while for GPT-4 via ChatGPT, the prompt with a Common Weakness Enumeration (CWE) list works best. (3) GPT-4 via ChatGPT frequently produces vague expressions and exhibits difficulties in accurately following instructions in the prompts, while DeepSeek-R1 more commonly generates inaccurate code details in its outputs. (4) LLMs are more adept at identifying security defects in code files that have fewer tokens and security-relevant annotations. (5) Higher code complexity correlates with enhanced detection capabilities of DeepSeek-R1 for specific security defect types.

cs.SE↗

arXiv2026

On Fixing Insecure AI-Generated Code through Model Fine-Tuning and Prompting Strategies

The security of AI-generated code remains a major obstacle to its widespread adoption. Although code generation models achieve strong performance on functional benchmarks, their outputs frequently contain bugs and security weaknesses that undermine their trustworthiness. Prior work has explored a range of approaches to mitigate security issues in AI-generated code, e.g., using static analysis-guided generation and prompt engineering. However, their effectiveness varies widely across models and settings. This paper presents a systematic investigation of strategies for hardening model-generated code against a list of Common Weakness Enumeration (CWE). We assess the extent to which these strategies improve security across models and programming languages, using fine-tuning and prompting approaches for model output refinement. Beyond the prevalence of security weaknesses, we analyse the severity of identified CWEs, their co-occurrence, and the unintended consequences of remediation (i.e., whether fixing certain weaknesses introduces new weaknesses elsewhere in the same code). Our results show that security improvements are highly strategy- and model-dependent. Although some approaches reduce specific classes of weaknesses, they often introduce new weaknesses as side effects of the fixes. Moreover, no strategy consistently eliminates weaknesses across all models and scenarios, highlighting the absence of a universally effective "bulletproof" solution for secure AI-generated code.

cs.SE↗

arXiv2026

When T-Depth Misleads: Predicting Fault-Tolerant Quantum Execution Slowdown under Magic-State Delivery Constraints

The efficient execution of fault-tolerant quantum algorithms is fundamentally limited by the production rate of magic states required for non-Clifford operations. While circuit optimization typically targets T-depth, static T-depth does not reliably predict executable performance under bounded T-state delivery. We introduce a model that captures demand-supply imbalance using two key quantities: slack ratio, a structural indicator of scheduling flexibility, and Delta_max, a measure of cumulative demand surplus. We show that Delta_max is a strong schedule-level indicator of execution slowdown and yields a provable lower bound on executable makespan for a fixed schedule. Empirical evaluation on constructed directed acyclic graph (DAG) families, with arithmetic circuits and exact quantum Fourier transform (QFT) traces providing additional grounding, shows that slack ratio is a stronger structural predictor than T-depth for stall and inversion risk, while Delta_max is the strongest predictor of slowdown. Across 4,904 instances, the lower bound shows zero violations, with 88.9% of cases within one cycle. These results highlight the importance of explicitly modeling delivery constraints in fault-tolerant quantum compilation.

quant-ph↗

arXiv2026

CoRaCMG: Contextual Retrieval-Augmented Framework for Commit Message Generation

Commit messages play a key role in documenting the intent behind code changes. However, they are often low-quality, vague, or incomplete, limiting their usefulness. Commit Message Generation (CMG) aims to automatically generate descriptive commit messages from code diffs to reduce developers' effort and improve message quality. Although recent advances in LLMs have shown promise in automating CMG, their performance remains limited. This paper aims to enhance CMG performance by retrieving similar diff-message pairs to guide LLMs to generate commit messages that are more precise and informative. We proposed CoRaCMG, a Contextual Retrieval-augmented framework for Commit Message Generation, structured in three phases: (1) Retrieve: retrieving the similar diff-message pairs; (2) Augment: combining them with the query diff into a structured prompt; and (3) Generate: generating commit messages corresponding to the query diff via LLMs. CoRaCMG enables LLMs to learn project-specific terminologies and writing styles from the retrieved diff-message pairs. We evaluated CoRaCMG across multiple LLMs (e.g., GPT, DeepSeek, and Qwen) and compared its performance against SOTA baselines. Experimental results show that CoRaCMG significantly boosts LLM performance across four metrics (BLEU, Rouge-L, METEOR, and CIDEr). Specifically, DeepSeek-R1 achieves relative improvements of 76% in BLEU and 71% in CIDEr when augmented with a single retrieved example pair. After incorporating the single example pair, GPT-4o achieves the highest improvement rate, with BLEU increasing by 89%. Moreover, performance gains plateau after more than three examples are used, indicating diminishing returns. Further analysis shows that the improvements are attributed to the model's ability to capture the terminologies and writing styles of human-written commit messages from the retrieved example pairs.

cs.SE↗

arXiv2026

Fine-Tuning Code Language Models to Detect Cross-Language Bugs

Multilingual programming, which involves using multiple programming languages (PLs) in a single project, is increasingly common due to its benefits. However, it introduces cross-language bugs (CLBs), which arise from interactions between different PLs and are difficult to detect by single-language bug detection tools. This paper investigates the potential of pre-trained code language models (CodeLMs) in CLB detection. We developed CLCFinder, a cross-language code identification tool, and constructed a CLB dataset involving three PL combinations (Python-C/C++, Java-C/C++, and Python-Java) with nine interaction types. We fine-tuned 13 CodeLMs on this dataset and evaluated their performance, analyzing the effects of dataset size, token sequence length, and code comments. Results show that all 13 CodeLMs exhibited varying degrees of performance improvement after fine-tuning, with UniXcoder-base achieving the best F1 score (0.7407). Notably, within our experimental setup, small CodeLMs tended to performe better than large ones. CodeLMs fine-tuned on single-language bug datasets performed poorly on CLB detection, demonstrating the distinction between CLBs and single-language bugs. Additionally, increasing the fine-tuning dataset size significantly improved performance, while longer token sequences did not necessarily improve the model performance. The impact of code comments varied across models. Some fine-tuned CodeLMs' performance was improved, while others showed degraded performance.

cs.SE↗

arXiv2026

Beyond Functional Correctness: Design Issues in AI IDE-Generated Large-Scale Projects

New generation of AI coding tools, including AI-powered IDEs equipped with agentic capabilities, can generate code within the context of the project. These AI IDEs are increasingly perceived as capable of producing project-level code at scale. However, there is limited empirical evidence on the extent to which they can generate large-scale software systems and what design issues such systems may exhibit. To address this gap, we conducted a study to explore the capability of Cursor in generating large-scale projects and to evaluate the design quality of projects generated by Cursor. First, we propose a Feature-Driven Human-In-The-Loop (FD-HITL) framework that systematically guides project generation from curated project descriptions. We generated 10 projects using Cursor with the FD-HITL framework across three application domains and multiple technologies. We assessed the functional correctness of these projects through manual evaluation, obtaining an average functional correctness score of 91%. Next, we analyzed the generated projects using two static analysis tools, CodeScene and SonarQube, to detect design issues. We identified 1,305 design issues categorized into 9 categories by CodeScene and 3,193 issues in 11 categories by SonarQube. Our findings show that (1) when used with the FD-HITL framework, Cursor can generate functional large-scale projects averaging 16,965 LoC and 114 files; (2) the generated projects nevertheless contain design issues that may pose long-term maintainability and evolvability risks, requiring careful review by experienced developers; (3) the most prevalent issues include Code Duplication, high Code Complexity, Large Methods, Framework Best-Practice Violations, Exception-Handling Issues and Accessibility Issues; (4) these design issues violate design principles such as SRP, SoC, and DRY. The replication package is at https://github.com/Kashifraz/DIinAGP

cs.SE↗

arXiv2026

Unveiling Code Clones in the Eclipse IIoT Software Ecosystem

Industrial Internet of Things (IIoT) has become a prominent topic recently, with an increasing number of IIoT OSS projects emerging, also within the Eclipse Foundation. Code cloning is a common practice that can adversely affect software maintenance. In the IIoT OSS domain, developers frequently reuse code and configurations for efficiency, which can lead to code clone proliferation and maintenance challenges. However, the extent and effects of code clones in the IIoT OSS domain remain understudied. This study aims to investigate the prevalence, evolution, and co-modification of code clones within the Eclipse IIoT OSS ecosystem. We collected 90 release versions from 15 projects in the Eclipse IIoT OSS ecosystem, and investigated their code clone situations based on source code and change history using the NiCad tool and our custom analysis module. The investigation covered clone distribution, patterns, evolution trends, co-modified clones, and cross-project clones. 1) Code clones are prevalent in Eclipse IIoT OSS projects, with 16.3% of code lines involved in clones - nearly twice the proportion observed in traditional OSS projects; 2) Most code clones occur between commits, while there are still a significant proportion of code clones that each clone pair happens within a commit; 3) Most Eclipse IIoT projects remain stable in clone numbers during version iterations; 4) An average of 0.17% of the clones have been co-modified, which negatively affect maintenance; and 5) Cross-project clone pairs are prevalent, more in Java than in C projects, with rare co-modifications (0.02%) only in Java projects. Our findings highlight the potential negative impacts of these clones on software maintenance, emphasizing the need to address these issues to improve overall software quality.

cs.SE↗

arXiv2026

A Comprehensive Evaluation of Parameter-Efficient Fine-Tuning on Code Smell Detection

Automated code smell detection faces persistent challenges due to the subjectivity of heuristic rules and the limited performance of traditional ML/DL models. While Large Language Models (LLMs) offer a promising alternative, their adoption is impeded by high fine-tuning costs and a lack of "LM-ready" benchmarks. To bridge these gaps, we present a study with two synergistic contributions. First, we constructed a high-quality benchmark for Complex Conditional, Complex Method, Feature Envy, and Data Class, validated through a rigorous two-stage manual review. Second, leveraging this benchmark, we systematically evaluated four Parameter-Efficient Fine-Tuning (PEFT) methods across nine LMs of varying parameter sizes. Their performance is compared against a comprehensive suite of baselines, including heuristics-based detectors, Deep Learning (DL)-based approaches, and state-of-the-art general-purpose LLMs under multiple In-Context Learning (ICL) settings. Our results demonstrate that PEFT methods achieve effectiveness comparable to or surpassing full fine-tuning while substantially reducing peak GPU memory usage for code smell detection. Furthermore, PEFT-tuned LMs consistently outperform all baselines, yielding MCC improvements ranging from 0.33% to 13.69%, with particularly notable gains for specific smell categories. These findings highlight PEFT techniques as effective and scalable solutions for advancing code smell detection.

cs.SE↗

arXiv2026

C2|Q>: A Robust Framework for Bridging Classical and Quantum Software Development

QSE is emerging as a critical discipline to make quantum computing accessible to a broader developer community; however, most quantum development environments still require developers to engage with low-level details across the software stack - including problem encoding, circuit construction, algorithm configuration, hardware selection, and result interpretation - making them difficult for classical software engineers to use. To bridge this gap, we present C2|Q>, a hardware-agnostic quantum software development framework that translates specific types of classical specifications into quantum-executable programs while preserving methodological rigor. The framework applies modular SE principles by classifying the workflow into three core modules: an encoder that classifies problems, produces Quantum-Compatible Formats, and constructs quantum circuits, a deployment module that generates circuits and recommends hardware based on fidelity, runtime, and cost, and a decoder that interprets quantum outputs into classical solutions. In evaluation, the encoder module achieved a 93.8% completion rate, the hardware recommendation module consistently selected the appropriate quantum devices for workloads scaling up to 56 qubits. End-to-end experiments on 434 Python programs and 100 JSON problem instances show that the full C2|Q> workflow executes reliably on simulators and can be deployed successfully on representative real quantum hardware, with empirical runs limited to small- and medium-sized instances consistent with current NISQ capabilities. These results indicate that C2|Q> lowers the entry barrier to quantum software development by providing a reproducible, extensible toolchain that connects classical specifications to quantum execution. The open-source implementation of C2|Q> is available at https://github.com/C2-Q/C2Q and as a Python package at https://pypi.org/project/c2q-framework/.

cs.SE↗

arXiv2026

Understanding the Issues, Their Causes and Solutions in Microservices Systems: An Empirical Study

Many small to large organizations have adopted the Microservices Architecture (MSA) style to develop and deliver their core businesses. Despite the popularity of MSA in the software industry, there is a limited evidence-based and thorough understanding of the types of issues (e.g., errors, faults, failures, and bugs) that microservices system developers experience, the causes of the issues, and the solutions as potential fixing strategies to address the issues. To ameliorate this gap, we conducted a mixed-methods empirical study that collected data from 2,641 issues from the issue tracking systems of 15 open-source microservices systems on GitHub, 15 interviews, and an online survey completed by 150 practitioners from 42 countries across 6 continents. Our analysis led to comprehensive taxonomies for the issues, causes, and solutions. The findings of this study informthat Technical Debt, Continuous Integration and Delivery, Exception Handling, Service Execution and Communication, and Security are the most dominant issues in microservices systems. Furthermore, General Programming Errors, Missing Features and Artifacts, and Invalid Configuration and Communication are the main causes behind the issues. Finally, we found 177 types of solutions that can be applied to fix the identified issues. Based on our study results, we propose a future research framework that outlines key problem dimensions and actionable study strategies to support the engineering of emergent and next-generation microservices systems.

cs.SE↗

arXiv2026

Unveiling the Role of ChatGPT in Software Development: Insights from Developer-ChatGPT Interactions on GitHub

The advent of Large Language Models (LLMs) has introduced a new paradigm in Software Engineering (SE), with generative AI tools like ChatGPT gaining widespread adoption among developers. While ChatGPT's potential has been extensively discussed, empirical evidence about how developers actually use LLMs' assistance in real-world practices remains limited. To bridge this gap, we conducted a large-scale empirical analysis of ChatGPT usage on GitHub, and we presented DevChat, a curated dataset of 2,547 publicly shared ChatGPT conversation links collected from GitHub between May 2023 and June 2024. Through comprehensively analyzing DevChat, we explored the characteristics of developer-ChatGPT interaction patterns and identified five key categories of developers' purposes for sharing developer-ChatGPT conversations during software development. Additionally, we investigated the dominant development-related activities in which ChatGPT is used, and presented a mapping framework that links GitHub data sources, development-related activities, and SE tasks. The findings show that interactions are typically short and task-focused (most are 1-3 turns); developers share conversations mainly to delegate tasks, resolve problems, and acquire knowledge, revealing five purpose categories; ChatGPT is most frequently engaged for Software Implementation and Maintenance & Evolution; we identified 39 fine-grained SE tasks supported by ChatGPT, with Code Generation & Completion as well as Code modification & Optimization being the most prominent. Our study offers a comprehensive mapping of ChatGPT's applications in real-world software development scenarios and provides a foundation for understanding LLMs' practical roles in software development.

cs.SE↗

arXiv2025

Bug Priority Change Prediction: An Exploratory Study on Apache Software

Bug fixing is a critical activity in the software development process. In issue tracking systems such as JIRA, each bug report is assigned a priority level to indicate the urgency and importance level of the bug. The priority may change during the bug fixing process, indicating that the urgency and importance level of the bug will change with the bug fixing. However, manually evaluating priority changes for bugs is a tedious process that heavily relies on the subjective judgment of developers and project managers, leading to incorrect priority changes and thus hindering timely bug fixes. Given the lack of research on bug priority change prediction, we propose a novel two-phase bug report priority change prediction method based on bug fixing evolution features and class imbalance handling strategy. Specifically, we divided the bug lifecycle into two phases: bug reporting and bug fixing, and constructed bug priority change prediction models for each phase. To evaluate the performance of our method, we conducted experiments on a bug dataset constructed from 32 non-trivial Apache projects. The experimental results show that our proposed bug fixing evolution features and the adopted class imbalance handling strategy can effectively improve the performance of prediction models. The F1-score of the prediction model constructed for the bug reporting phase reached 0.798, while the F1-weighted and F1-macro of the prediction model constructed for the bug fixing phase were 0.712 and 0.613, respectively. Furthermore, we explored the cross-project applicability of our prediction models and their performance at different priority levels. The findings indicate large variations in model performance across different projects, although the overall scores remain decent. Meanwhile, the predictive performance across various priority levels remained relatively consistently high.

cs.SE↗

arXiv2025

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. Additionally, 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.

cs.SE↗

arXiv2025

Designing LLM-based Multi-Agent Systems for Software Engineering Tasks: Quality Attributes, Design Patterns and Rationale

As the complexity of Software Engineering (SE) tasks continues to escalate, Multi-Agent Systems (MASs) have emerged as a focal point of research and practice due to their autonomy and scalability. Furthermore, through leveraging the reasoning and planning capabilities of Large Language Models (LLMs), the application of LLM-based MASs in the field of SE is garnering increasing attention. However, there is no dedicated study that systematically explores the design of LLM-based MASs, including the Quality Attributes (QAs) on which designers mainly focus, the design patterns used by designers, and the rationale guiding the design of LLM-based MASs for SE tasks. To this end, we conducted a study to identify the QAs that LLM-based MASs for SE tasks focus on, the design patterns used in the MASs, and the design rationale for the MASs. We collected 94 papers on LLM-based MASs for SE tasks as the source. Our study shows that: (1) Code Generation is the most common SE task solved by LLM-based MASs among ten identified SE tasks, (2) Functional Suitability is the QA on which designers of LLM-based MASs pay the most attention, (3) Role-Based Cooperation is the design pattern most frequently employed among 16 patterns used to construct LLM-based MASs, and (4) Improving the Quality of Generated Code is the most common rationale behind the design of LLM-based MASs. Based on the study results, we presented the implications for the design of LLM-based MASs to support SE tasks.

cs.SE↗

arXiv2025

A Survey of Bugs in AI-Generated Code

Developers are widely using AI code-generation models, aiming to increase productivity and efficiency. However, there are also quality concerns regarding the AI-generated code. The generated code is produced by models trained on publicly available code, which are known to contain bugs and quality issues. Those issues can cause trust and maintenance challenges during the development process. Several quality issues associated with AI-generated code have been reported, including bugs and defects. However, these findings are often scattered and lack a systematic summary. A comprehensive review is currently lacking to reveal the types and distribution of these errors, possible remediation strategies, as well as their correlation with the specific models. In this paper, we systematically analyze the existing AI-generated code literature to establish an overall understanding of bugs and defects in generated code, providing a reference for future model improvement and quality assessment. We aim to understand the nature and extent of bugs in AI-generated code, and provide a classification of bug types and patterns present in code generated by different models. We also discuss possible fixes and mitigation strategies adopted to eliminate bugs from the generated code.

cs.SE↗